diff --git "a/sft/smoe_perturbed/trainer_state.json" "b/sft/smoe_perturbed/trainer_state.json" new file mode 100644--- /dev/null +++ "b/sft/smoe_perturbed/trainer_state.json" @@ -0,0 +1,58254 @@ +{ + "best_metric": null, + "best_model_checkpoint": null, + "epoch": 0.9999398785546805, + "eval_steps": 500, + "global_step": 8316, + "is_hyper_param_search": false, + "is_local_process_zero": true, + "is_world_process_zero": true, + "log_history": [ + { + "epoch": 0.00012024289063909097, + "grad_norm": 39.989183107748026, + "learning_rate": 0.0, + "loss": 2.0936, + "step": 1 + }, + { + "epoch": 0.00024048578127818193, + "grad_norm": 61.49701664829461, + "learning_rate": 5.021476677069823e-07, + "loss": 2.1277, + "step": 2 + }, + { + "epoch": 0.0003607286719172729, + "grad_norm": 102.18368207699554, + "learning_rate": 7.958852231401551e-07, + "loss": 1.854, + "step": 3 + }, + { + "epoch": 0.00048097156255636386, + "grad_norm": 36.4769799786026, + "learning_rate": 1.0042953354139647e-06, + "loss": 1.8847, + "step": 4 + }, + { + "epoch": 0.0006012144531954548, + "grad_norm": 54.90794409025665, + "learning_rate": 1.1659507774310057e-06, + "loss": 2.1737, + "step": 5 + }, + { + "epoch": 0.0007214573438345458, + "grad_norm": 46.75867597886915, + "learning_rate": 1.2980328908471373e-06, + "loss": 1.8493, + "step": 6 + }, + { + "epoch": 0.0008417002344736367, + "grad_norm": 4.574437658217239, + "learning_rate": 1.4097067265369432e-06, + "loss": 1.0527, + "step": 7 + }, + { + "epoch": 0.0009619431251127277, + "grad_norm": 42.04976503213558, + "learning_rate": 1.506443003120947e-06, + "loss": 1.8331, + "step": 8 + }, + { + "epoch": 0.0010821860157518186, + "grad_norm": 18.15673575350636, + "learning_rate": 1.5917704462803102e-06, + "loss": 1.7185, + "step": 9 + }, + { + "epoch": 0.0012024289063909096, + "grad_norm": 13.455567318776165, + "learning_rate": 1.6680984451379884e-06, + "loss": 1.7716, + "step": 10 + }, + { + "epoch": 0.0013226717970300007, + "grad_norm": 14.120337873096048, + "learning_rate": 1.7371455188905097e-06, + "loss": 1.5655, + "step": 11 + }, + { + "epoch": 0.0014429146876690916, + "grad_norm": 10.668465780484793, + "learning_rate": 1.8001805585541196e-06, + "loss": 1.4918, + "step": 12 + }, + { + "epoch": 0.0015631575783081825, + "grad_norm": 6.824972232301, + "learning_rate": 1.8581671739548328e-06, + "loss": 1.5367, + "step": 13 + }, + { + "epoch": 0.0016834004689472734, + "grad_norm": 7.372974958258533, + "learning_rate": 1.9118543942439254e-06, + "loss": 1.3743, + "step": 14 + }, + { + "epoch": 0.0018036433595863645, + "grad_norm": 5.8457935044591975, + "learning_rate": 1.961836000571161e-06, + "loss": 1.366, + "step": 15 + }, + { + "epoch": 0.0019238862502254555, + "grad_norm": 3.7963530256241995, + "learning_rate": 2.0085906708279293e-06, + "loss": 0.8851, + "step": 16 + }, + { + "epoch": 0.0020441291408645466, + "grad_norm": 5.346265412746115, + "learning_rate": 2.0525099325728135e-06, + "loss": 1.3995, + "step": 17 + }, + { + "epoch": 0.0021643720315036373, + "grad_norm": 3.512364210443125, + "learning_rate": 2.0939181139872922e-06, + "loss": 0.9598, + "step": 18 + }, + { + "epoch": 0.0022846149221427284, + "grad_norm": 4.8693190510203666, + "learning_rate": 2.1330868934640175e-06, + "loss": 1.2534, + "step": 19 + }, + { + "epoch": 0.002404857812781819, + "grad_norm": 3.572882530709213, + "learning_rate": 2.170246112844971e-06, + "loss": 1.0038, + "step": 20 + }, + { + "epoch": 0.0025251007034209102, + "grad_norm": 5.242601904144715, + "learning_rate": 2.2055919496770983e-06, + "loss": 1.2548, + "step": 21 + }, + { + "epoch": 0.0026453435940600014, + "grad_norm": 4.483816504175348, + "learning_rate": 2.2392931865974923e-06, + "loss": 1.1332, + "step": 22 + }, + { + "epoch": 0.002765586484699092, + "grad_norm": 4.393882547606739, + "learning_rate": 2.271496085962064e-06, + "loss": 1.2573, + "step": 23 + }, + { + "epoch": 0.002885829375338183, + "grad_norm": 2.9457903810757275, + "learning_rate": 2.3023282262611022e-06, + "loss": 1.2649, + "step": 24 + }, + { + "epoch": 0.003006072265977274, + "grad_norm": 3.371244672488882, + "learning_rate": 2.3319015548620114e-06, + "loss": 1.1636, + "step": 25 + }, + { + "epoch": 0.003126315156616365, + "grad_norm": 2.427941679447348, + "learning_rate": 2.3603148416618152e-06, + "loss": 1.1676, + "step": 26 + }, + { + "epoch": 0.003246558047255456, + "grad_norm": 2.4459733648900817, + "learning_rate": 2.3876556694204647e-06, + "loss": 1.2494, + "step": 27 + }, + { + "epoch": 0.003366800937894547, + "grad_norm": 3.0931321885493097, + "learning_rate": 2.414002061950908e-06, + "loss": 1.1441, + "step": 28 + }, + { + "epoch": 0.003487043828533638, + "grad_norm": 4.661318689169093, + "learning_rate": 2.4394238264681557e-06, + "loss": 1.238, + "step": 29 + }, + { + "epoch": 0.003607286719172729, + "grad_norm": 2.0332265773743483, + "learning_rate": 2.4639836682781433e-06, + "loss": 1.2345, + "step": 30 + }, + { + "epoch": 0.00372752960981182, + "grad_norm": 3.310180571779377, + "learning_rate": 2.487738122623307e-06, + "loss": 1.2409, + "step": 31 + }, + { + "epoch": 0.003847772500450911, + "grad_norm": 2.475551266399224, + "learning_rate": 2.510738338534912e-06, + "loss": 1.227, + "step": 32 + }, + { + "epoch": 0.003968015391090002, + "grad_norm": 2.3846723458607553, + "learning_rate": 2.5330307420306648e-06, + "loss": 1.2636, + "step": 33 + }, + { + "epoch": 0.004088258281729093, + "grad_norm": 2.399522666351929, + "learning_rate": 2.554657600279796e-06, + "loss": 1.1199, + "step": 34 + }, + { + "epoch": 0.004208501172368184, + "grad_norm": 2.1341194368454253, + "learning_rate": 2.5756575039679493e-06, + "loss": 1.2743, + "step": 35 + }, + { + "epoch": 0.0043287440630072746, + "grad_norm": 2.057043776888723, + "learning_rate": 2.5960657816942747e-06, + "loss": 1.1885, + "step": 36 + }, + { + "epoch": 0.004448986953646365, + "grad_norm": 1.3422483881150087, + "learning_rate": 2.6159148575788668e-06, + "loss": 0.8479, + "step": 37 + }, + { + "epoch": 0.004569229844285457, + "grad_norm": 3.2884017751859926, + "learning_rate": 2.635234561171e-06, + "loss": 1.2241, + "step": 38 + }, + { + "epoch": 0.0046894727349245475, + "grad_norm": 2.2744677329702334, + "learning_rate": 2.6540523970949877e-06, + "loss": 1.1799, + "step": 39 + }, + { + "epoch": 0.004809715625563638, + "grad_norm": 3.115501158002535, + "learning_rate": 2.6723937805519533e-06, + "loss": 1.1641, + "step": 40 + }, + { + "epoch": 0.00492995851620273, + "grad_norm": 2.3742966865383077, + "learning_rate": 2.690282243737839e-06, + "loss": 1.1681, + "step": 41 + }, + { + "epoch": 0.0050502014068418205, + "grad_norm": 3.547226270850757, + "learning_rate": 2.7077396173840807e-06, + "loss": 1.2323, + "step": 42 + }, + { + "epoch": 0.005170444297480911, + "grad_norm": 2.417705663317699, + "learning_rate": 2.7247861909342594e-06, + "loss": 1.1649, + "step": 43 + }, + { + "epoch": 0.005290687188120003, + "grad_norm": 2.428966550179745, + "learning_rate": 2.7414408543044743e-06, + "loss": 1.0679, + "step": 44 + }, + { + "epoch": 0.005410930078759093, + "grad_norm": 4.53156126860577, + "learning_rate": 2.7577212237113157e-06, + "loss": 1.0307, + "step": 45 + }, + { + "epoch": 0.005531172969398184, + "grad_norm": 2.0449917115507534, + "learning_rate": 2.7736437536690466e-06, + "loss": 1.2799, + "step": 46 + }, + { + "epoch": 0.005651415860037276, + "grad_norm": 2.1186638783644005, + "learning_rate": 2.789223836941131e-06, + "loss": 1.3155, + "step": 47 + }, + { + "epoch": 0.005771658750676366, + "grad_norm": 2.325157886189501, + "learning_rate": 2.8044758939680847e-06, + "loss": 1.3246, + "step": 48 + }, + { + "epoch": 0.005891901641315457, + "grad_norm": 2.7225558072229856, + "learning_rate": 2.8194134530738863e-06, + "loss": 1.2593, + "step": 49 + }, + { + "epoch": 0.006012144531954548, + "grad_norm": 2.604334548625063, + "learning_rate": 2.834049222568994e-06, + "loss": 1.1407, + "step": 50 + }, + { + "epoch": 0.006132387422593639, + "grad_norm": 2.02649778381874, + "learning_rate": 2.848395155712969e-06, + "loss": 1.163, + "step": 51 + }, + { + "epoch": 0.00625263031323273, + "grad_norm": 2.2544023043948997, + "learning_rate": 2.8624625093687977e-06, + "loss": 1.2159, + "step": 52 + }, + { + "epoch": 0.006372873203871821, + "grad_norm": 2.480012626461988, + "learning_rate": 2.876261897070029e-06, + "loss": 1.1273, + "step": 53 + }, + { + "epoch": 0.006493116094510912, + "grad_norm": 2.7795252220298168, + "learning_rate": 2.889803337127447e-06, + "loss": 1.1616, + "step": 54 + }, + { + "epoch": 0.006613358985150003, + "grad_norm": 2.476449765307989, + "learning_rate": 2.903096296321516e-06, + "loss": 1.0853, + "step": 55 + }, + { + "epoch": 0.006733601875789094, + "grad_norm": 2.2241386203514972, + "learning_rate": 2.9161497296578907e-06, + "loss": 1.1541, + "step": 56 + }, + { + "epoch": 0.006853844766428185, + "grad_norm": 2.52708506844868, + "learning_rate": 2.928972116604173e-06, + "loss": 1.0956, + "step": 57 + }, + { + "epoch": 0.006974087657067276, + "grad_norm": 2.3799633565053275, + "learning_rate": 2.9415714941751377e-06, + "loss": 1.2576, + "step": 58 + }, + { + "epoch": 0.007094330547706367, + "grad_norm": 2.1793514525900246, + "learning_rate": 2.9539554871897396e-06, + "loss": 1.1741, + "step": 59 + }, + { + "epoch": 0.007214573438345458, + "grad_norm": 2.000870872673944, + "learning_rate": 2.9661313359851253e-06, + "loss": 1.2112, + "step": 60 + }, + { + "epoch": 0.007334816328984549, + "grad_norm": 2.3931280259327203, + "learning_rate": 2.978105921839922e-06, + "loss": 1.1759, + "step": 61 + }, + { + "epoch": 0.00745505921962364, + "grad_norm": 2.3914175278186782, + "learning_rate": 2.9898857903302893e-06, + "loss": 0.9604, + "step": 62 + }, + { + "epoch": 0.007575302110262731, + "grad_norm": 3.042070092050196, + "learning_rate": 3.001477172817253e-06, + "loss": 1.1166, + "step": 63 + }, + { + "epoch": 0.007695545000901822, + "grad_norm": 5.660119519269192, + "learning_rate": 3.012886006241894e-06, + "loss": 1.1998, + "step": 64 + }, + { + "epoch": 0.007815787891540913, + "grad_norm": 2.15535275132297, + "learning_rate": 3.0241179513858383e-06, + "loss": 1.119, + "step": 65 + }, + { + "epoch": 0.007936030782180003, + "grad_norm": 2.194351474654329, + "learning_rate": 3.035178409737647e-06, + "loss": 1.1157, + "step": 66 + }, + { + "epoch": 0.008056273672819095, + "grad_norm": 2.4399444842667446, + "learning_rate": 3.046072539090907e-06, + "loss": 1.1252, + "step": 67 + }, + { + "epoch": 0.008176516563458186, + "grad_norm": 2.827495357330855, + "learning_rate": 3.056805267986779e-06, + "loss": 1.2867, + "step": 68 + }, + { + "epoch": 0.008296759454097276, + "grad_norm": 2.254412245520749, + "learning_rate": 3.0673813091022194e-06, + "loss": 1.1915, + "step": 69 + }, + { + "epoch": 0.008417002344736368, + "grad_norm": 1.3357231905183564, + "learning_rate": 3.0778051716749317e-06, + "loss": 0.8588, + "step": 70 + }, + { + "epoch": 0.008537245235375458, + "grad_norm": 2.777913353048633, + "learning_rate": 3.0880811730470094e-06, + "loss": 1.141, + "step": 71 + }, + { + "epoch": 0.008657488126014549, + "grad_norm": 1.1369382246576205, + "learning_rate": 3.098213449401257e-06, + "loss": 0.824, + "step": 72 + }, + { + "epoch": 0.00877773101665364, + "grad_norm": 2.0964008460366506, + "learning_rate": 3.1082059657570015e-06, + "loss": 1.2264, + "step": 73 + }, + { + "epoch": 0.00889797390729273, + "grad_norm": 3.601762679157617, + "learning_rate": 3.1180625252858496e-06, + "loss": 1.2041, + "step": 74 + }, + { + "epoch": 0.009018216797931822, + "grad_norm": 2.654185491362387, + "learning_rate": 3.1277867780021663e-06, + "loss": 1.0382, + "step": 75 + }, + { + "epoch": 0.009138459688570914, + "grad_norm": 2.031695742483814, + "learning_rate": 3.1373822288779824e-06, + "loss": 1.1956, + "step": 76 + }, + { + "epoch": 0.009258702579210003, + "grad_norm": 1.9290168616159584, + "learning_rate": 3.1468522454274533e-06, + "loss": 1.0331, + "step": 77 + }, + { + "epoch": 0.009378945469849095, + "grad_norm": 2.0574007287069493, + "learning_rate": 3.15620006480197e-06, + "loss": 1.1548, + "step": 78 + }, + { + "epoch": 0.009499188360488187, + "grad_norm": 3.3848109872357015, + "learning_rate": 3.1654288004333087e-06, + "loss": 0.9853, + "step": 79 + }, + { + "epoch": 0.009619431251127276, + "grad_norm": 2.631456880750012, + "learning_rate": 3.1745414482589353e-06, + "loss": 0.998, + "step": 80 + }, + { + "epoch": 0.009739674141766368, + "grad_norm": 3.089758870614346, + "learning_rate": 3.1835408925606204e-06, + "loss": 1.1067, + "step": 81 + }, + { + "epoch": 0.00985991703240546, + "grad_norm": 2.23071319790496, + "learning_rate": 3.1924299114448214e-06, + "loss": 1.1304, + "step": 82 + }, + { + "epoch": 0.00998015992304455, + "grad_norm": 2.0494452493487416, + "learning_rate": 3.2012111819909055e-06, + "loss": 1.0705, + "step": 83 + }, + { + "epoch": 0.010100402813683641, + "grad_norm": 2.408403201118221, + "learning_rate": 3.2098872850910627e-06, + "loss": 1.1876, + "step": 84 + }, + { + "epoch": 0.010220645704322733, + "grad_norm": 1.96501738310058, + "learning_rate": 3.2184607100038194e-06, + "loss": 1.1287, + "step": 85 + }, + { + "epoch": 0.010340888594961822, + "grad_norm": 3.821168518894774, + "learning_rate": 3.2269338586412414e-06, + "loss": 1.1684, + "step": 86 + }, + { + "epoch": 0.010461131485600914, + "grad_norm": 2.17244367221377, + "learning_rate": 3.2353090496083106e-06, + "loss": 1.2015, + "step": 87 + }, + { + "epoch": 0.010581374376240005, + "grad_norm": 1.9343813746629237, + "learning_rate": 3.2435885220114572e-06, + "loss": 1.0513, + "step": 88 + }, + { + "epoch": 0.010701617266879095, + "grad_norm": 2.5838699603826756, + "learning_rate": 3.2517744390519113e-06, + "loss": 1.1777, + "step": 89 + }, + { + "epoch": 0.010821860157518187, + "grad_norm": 4.657308778139114, + "learning_rate": 3.259868891418298e-06, + "loss": 0.9873, + "step": 90 + }, + { + "epoch": 0.010942103048157278, + "grad_norm": 3.953733254179224, + "learning_rate": 3.2678739004917757e-06, + "loss": 1.0873, + "step": 91 + }, + { + "epoch": 0.011062345938796368, + "grad_norm": 1.8600238982070216, + "learning_rate": 3.275791421376029e-06, + "loss": 1.1592, + "step": 92 + }, + { + "epoch": 0.01118258882943546, + "grad_norm": 2.051008055724603, + "learning_rate": 3.2836233457634622e-06, + "loss": 1.2006, + "step": 93 + }, + { + "epoch": 0.011302831720074551, + "grad_norm": 1.971855015963266, + "learning_rate": 3.2913715046481135e-06, + "loss": 1.0926, + "step": 94 + }, + { + "epoch": 0.011423074610713641, + "grad_norm": 2.10370331248328, + "learning_rate": 3.299037670895023e-06, + "loss": 1.1271, + "step": 95 + }, + { + "epoch": 0.011543317501352733, + "grad_norm": 2.504399719649407, + "learning_rate": 3.3066235616750667e-06, + "loss": 1.0407, + "step": 96 + }, + { + "epoch": 0.011663560391991824, + "grad_norm": 4.113007423147472, + "learning_rate": 3.3141308407736276e-06, + "loss": 1.1612, + "step": 97 + }, + { + "epoch": 0.011783803282630914, + "grad_norm": 2.677635986835557, + "learning_rate": 3.321561120780869e-06, + "loss": 1.105, + "step": 98 + }, + { + "epoch": 0.011904046173270006, + "grad_norm": 2.4215772876492285, + "learning_rate": 3.3289159651708192e-06, + "loss": 1.2526, + "step": 99 + }, + { + "epoch": 0.012024289063909096, + "grad_norm": 2.0479520337403465, + "learning_rate": 3.3361968902759768e-06, + "loss": 1.2156, + "step": 100 + }, + { + "epoch": 0.012144531954548187, + "grad_norm": 2.1608928507552307, + "learning_rate": 3.343405367163663e-06, + "loss": 1.1762, + "step": 101 + }, + { + "epoch": 0.012264774845187279, + "grad_norm": 6.23178865329497, + "learning_rate": 3.350542823419951e-06, + "loss": 1.0504, + "step": 102 + }, + { + "epoch": 0.012385017735826368, + "grad_norm": 3.3853290879096622, + "learning_rate": 3.3576106448465615e-06, + "loss": 1.1136, + "step": 103 + }, + { + "epoch": 0.01250526062646546, + "grad_norm": 2.192708925321111, + "learning_rate": 3.3646101770757797e-06, + "loss": 1.1204, + "step": 104 + }, + { + "epoch": 0.012625503517104552, + "grad_norm": 2.8697127777352174, + "learning_rate": 3.371542727108104e-06, + "loss": 1.0955, + "step": 105 + }, + { + "epoch": 0.012745746407743641, + "grad_norm": 2.8767330088061835, + "learning_rate": 3.3784095647770114e-06, + "loss": 1.1371, + "step": 106 + }, + { + "epoch": 0.012865989298382733, + "grad_norm": 27.696575253300157, + "learning_rate": 3.3852119241449547e-06, + "loss": 1.1239, + "step": 107 + }, + { + "epoch": 0.012986232189021825, + "grad_norm": 2.5495663330222875, + "learning_rate": 3.3919510048344295e-06, + "loss": 1.2, + "step": 108 + }, + { + "epoch": 0.013106475079660914, + "grad_norm": 2.8196334232665, + "learning_rate": 3.3986279732976907e-06, + "loss": 1.1058, + "step": 109 + }, + { + "epoch": 0.013226717970300006, + "grad_norm": 2.0112078860599434, + "learning_rate": 3.4052439640284983e-06, + "loss": 1.1917, + "step": 110 + }, + { + "epoch": 0.013346960860939098, + "grad_norm": 1.8898453799584098, + "learning_rate": 3.4118000807190217e-06, + "loss": 1.0486, + "step": 111 + }, + { + "epoch": 0.013467203751578187, + "grad_norm": 1.7482478361919622, + "learning_rate": 3.4182973973648723e-06, + "loss": 0.9987, + "step": 112 + }, + { + "epoch": 0.013587446642217279, + "grad_norm": 2.667598938899159, + "learning_rate": 3.424736959321014e-06, + "loss": 1.1903, + "step": 113 + }, + { + "epoch": 0.01370768953285637, + "grad_norm": 1.9783941534774288, + "learning_rate": 3.431119784311155e-06, + "loss": 1.1237, + "step": 114 + }, + { + "epoch": 0.01382793242349546, + "grad_norm": 1.8472702558705039, + "learning_rate": 3.43744686339307e-06, + "loss": 1.0147, + "step": 115 + }, + { + "epoch": 0.013948175314134552, + "grad_norm": 2.162796476470408, + "learning_rate": 3.44371916188212e-06, + "loss": 1.1469, + "step": 116 + }, + { + "epoch": 0.014068418204773643, + "grad_norm": 1.953887729488363, + "learning_rate": 3.449937620235143e-06, + "loss": 1.0988, + "step": 117 + }, + { + "epoch": 0.014188661095412733, + "grad_norm": 1.6716835411465032, + "learning_rate": 3.456103154896722e-06, + "loss": 1.1308, + "step": 118 + }, + { + "epoch": 0.014308903986051825, + "grad_norm": 2.0740369009415636, + "learning_rate": 3.462216659109757e-06, + "loss": 1.163, + "step": 119 + }, + { + "epoch": 0.014429146876690916, + "grad_norm": 2.3597174389647058, + "learning_rate": 3.4682790036921077e-06, + "loss": 1.0898, + "step": 120 + }, + { + "epoch": 0.014549389767330006, + "grad_norm": 1.7874540069638458, + "learning_rate": 3.4742910377810193e-06, + "loss": 1.0691, + "step": 121 + }, + { + "epoch": 0.014669632657969098, + "grad_norm": 2.5530909411097484, + "learning_rate": 3.4802535895469042e-06, + "loss": 1.126, + "step": 122 + }, + { + "epoch": 0.01478987554860819, + "grad_norm": 6.713685011732943, + "learning_rate": 3.4861674668779934e-06, + "loss": 1.1356, + "step": 123 + }, + { + "epoch": 0.01491011843924728, + "grad_norm": 2.132389124395014, + "learning_rate": 3.492033458037272e-06, + "loss": 1.0793, + "step": 124 + }, + { + "epoch": 0.01503036132988637, + "grad_norm": 2.4978485386593237, + "learning_rate": 3.497852332293018e-06, + "loss": 1.105, + "step": 125 + }, + { + "epoch": 0.015150604220525462, + "grad_norm": 2.390793937747667, + "learning_rate": 3.5036248405242356e-06, + "loss": 1.2038, + "step": 126 + }, + { + "epoch": 0.015270847111164552, + "grad_norm": 1.987441749438891, + "learning_rate": 3.509351715802146e-06, + "loss": 1.0658, + "step": 127 + }, + { + "epoch": 0.015391090001803644, + "grad_norm": 2.7967392872538444, + "learning_rate": 3.5150336739488763e-06, + "loss": 1.0228, + "step": 128 + }, + { + "epoch": 0.015511332892442733, + "grad_norm": 2.011583401469324, + "learning_rate": 3.5206714140744143e-06, + "loss": 1.079, + "step": 129 + }, + { + "epoch": 0.015631575783081827, + "grad_norm": 5.8665392399445, + "learning_rate": 3.5262656190928208e-06, + "loss": 1.1138, + "step": 130 + }, + { + "epoch": 0.015751818673720917, + "grad_norm": 1.0438433348446428, + "learning_rate": 3.5318169562186737e-06, + "loss": 0.9535, + "step": 131 + }, + { + "epoch": 0.015872061564360006, + "grad_norm": 2.3584402898072145, + "learning_rate": 3.5373260774446292e-06, + "loss": 1.0611, + "step": 132 + }, + { + "epoch": 0.0159923044549991, + "grad_norm": 1.7846676347682655, + "learning_rate": 3.542793620000961e-06, + "loss": 1.1423, + "step": 133 + }, + { + "epoch": 0.01611254734563819, + "grad_norm": 2.077506525463226, + "learning_rate": 3.5482202067978894e-06, + "loss": 1.1084, + "step": 134 + }, + { + "epoch": 0.01623279023627728, + "grad_norm": 2.004821806145312, + "learning_rate": 3.553606446851471e-06, + "loss": 0.999, + "step": 135 + }, + { + "epoch": 0.016353033126916373, + "grad_norm": 12.929355877544833, + "learning_rate": 3.5589529356937613e-06, + "loss": 1.0712, + "step": 136 + }, + { + "epoch": 0.016473276017555463, + "grad_norm": 1.7369402223841395, + "learning_rate": 3.5642602557679627e-06, + "loss": 1.0076, + "step": 137 + }, + { + "epoch": 0.016593518908194552, + "grad_norm": 2.1844491664972985, + "learning_rate": 3.569528976809202e-06, + "loss": 1.0815, + "step": 138 + }, + { + "epoch": 0.016713761798833646, + "grad_norm": 1.792731774321146, + "learning_rate": 3.5747596562115522e-06, + "loss": 1.139, + "step": 139 + }, + { + "epoch": 0.016834004689472735, + "grad_norm": 2.3171564595258483, + "learning_rate": 3.5799528393819138e-06, + "loss": 1.1456, + "step": 140 + }, + { + "epoch": 0.016954247580111825, + "grad_norm": 4.405688094435331, + "learning_rate": 3.585109060081286e-06, + "loss": 1.1184, + "step": 141 + }, + { + "epoch": 0.017074490470750915, + "grad_norm": 3.05428130041831, + "learning_rate": 3.590228840753992e-06, + "loss": 1.0227, + "step": 142 + }, + { + "epoch": 0.01719473336139001, + "grad_norm": 2.619727607282605, + "learning_rate": 3.5953126928453423e-06, + "loss": 1.1119, + "step": 143 + }, + { + "epoch": 0.017314976252029098, + "grad_norm": 2.107668376997729, + "learning_rate": 3.600361117108239e-06, + "loss": 1.0436, + "step": 144 + }, + { + "epoch": 0.017435219142668188, + "grad_norm": 7.904151125262366, + "learning_rate": 3.6053746038991616e-06, + "loss": 1.2104, + "step": 145 + }, + { + "epoch": 0.01755546203330728, + "grad_norm": 1.0507735922155035, + "learning_rate": 3.6103536334639843e-06, + "loss": 0.8222, + "step": 146 + }, + { + "epoch": 0.01767570492394637, + "grad_norm": 2.735265968124825, + "learning_rate": 3.615298676214041e-06, + "loss": 1.0946, + "step": 147 + }, + { + "epoch": 0.01779594781458546, + "grad_norm": 2.6790629290188055, + "learning_rate": 3.6202101929928317e-06, + "loss": 1.129, + "step": 148 + }, + { + "epoch": 0.017916190705224554, + "grad_norm": 3.369839856109301, + "learning_rate": 3.6250886353337413e-06, + "loss": 1.1228, + "step": 149 + }, + { + "epoch": 0.018036433595863644, + "grad_norm": 1.9676403975652816, + "learning_rate": 3.6299344457091488e-06, + "loss": 1.1058, + "step": 150 + }, + { + "epoch": 0.018156676486502734, + "grad_norm": 2.095489007511857, + "learning_rate": 3.634748057771256e-06, + "loss": 1.1778, + "step": 151 + }, + { + "epoch": 0.018276919377141827, + "grad_norm": 1.7451916440205444, + "learning_rate": 3.639529896584965e-06, + "loss": 1.0959, + "step": 152 + }, + { + "epoch": 0.018397162267780917, + "grad_norm": 3.943005300782888, + "learning_rate": 3.6442803788531233e-06, + "loss": 1.126, + "step": 153 + }, + { + "epoch": 0.018517405158420007, + "grad_norm": 2.4513282756378034, + "learning_rate": 3.6489999131344357e-06, + "loss": 1.1989, + "step": 154 + }, + { + "epoch": 0.0186376480490591, + "grad_norm": 1.9263030732187414, + "learning_rate": 3.653688900054313e-06, + "loss": 1.1448, + "step": 155 + }, + { + "epoch": 0.01875789093969819, + "grad_norm": 1.943702123119798, + "learning_rate": 3.6583477325089526e-06, + "loss": 0.9984, + "step": 156 + }, + { + "epoch": 0.01887813383033728, + "grad_norm": 2.740046260192628, + "learning_rate": 3.6629767958628916e-06, + "loss": 1.2812, + "step": 157 + }, + { + "epoch": 0.018998376720976373, + "grad_norm": 3.3061400483567156, + "learning_rate": 3.667576468140291e-06, + "loss": 1.0938, + "step": 158 + }, + { + "epoch": 0.019118619611615463, + "grad_norm": 2.6708132770358572, + "learning_rate": 3.672147120210184e-06, + "loss": 1.1269, + "step": 159 + }, + { + "epoch": 0.019238862502254553, + "grad_norm": 1.939282059208156, + "learning_rate": 3.6766891159659177e-06, + "loss": 1.1013, + "step": 160 + }, + { + "epoch": 0.019359105392893646, + "grad_norm": 4.999478503687949, + "learning_rate": 3.6812028124990075e-06, + "loss": 1.1141, + "step": 161 + }, + { + "epoch": 0.019479348283532736, + "grad_norm": 7.05023424557514, + "learning_rate": 3.6856885602676016e-06, + "loss": 1.055, + "step": 162 + }, + { + "epoch": 0.019599591174171826, + "grad_norm": 2.0894288979304454, + "learning_rate": 3.6901467032597733e-06, + "loss": 1.1811, + "step": 163 + }, + { + "epoch": 0.01971983406481092, + "grad_norm": 7.448243916042313, + "learning_rate": 3.694577579151804e-06, + "loss": 1.1101, + "step": 164 + }, + { + "epoch": 0.01984007695545001, + "grad_norm": 2.33001284256044, + "learning_rate": 3.6989815194616703e-06, + "loss": 0.9762, + "step": 165 + }, + { + "epoch": 0.0199603198460891, + "grad_norm": 2.503438777997066, + "learning_rate": 3.703358849697888e-06, + "loss": 1.0347, + "step": 166 + }, + { + "epoch": 0.020080562736728192, + "grad_norm": 5.167959707746719, + "learning_rate": 3.7077098895038803e-06, + "loss": 1.0669, + "step": 167 + }, + { + "epoch": 0.020200805627367282, + "grad_norm": 4.2743423312963955, + "learning_rate": 3.712034952798045e-06, + "loss": 1.205, + "step": 168 + }, + { + "epoch": 0.02032104851800637, + "grad_norm": 2.1299152747841377, + "learning_rate": 3.7163343479096656e-06, + "loss": 1.0825, + "step": 169 + }, + { + "epoch": 0.020441291408645465, + "grad_norm": 2.5193284537147065, + "learning_rate": 3.720608377710802e-06, + "loss": 1.0699, + "step": 170 + }, + { + "epoch": 0.020561534299284555, + "grad_norm": 2.339830401705717, + "learning_rate": 3.7248573397443277e-06, + "loss": 1.1033, + "step": 171 + }, + { + "epoch": 0.020681777189923645, + "grad_norm": 1.9397605506363231, + "learning_rate": 3.729081526348224e-06, + "loss": 1.2127, + "step": 172 + }, + { + "epoch": 0.020802020080562738, + "grad_norm": 1.797718148821594, + "learning_rate": 3.7332812247762777e-06, + "loss": 1.0888, + "step": 173 + }, + { + "epoch": 0.020922262971201828, + "grad_norm": 2.469373213898426, + "learning_rate": 3.737456717315293e-06, + "loss": 1.1937, + "step": 174 + }, + { + "epoch": 0.021042505861840918, + "grad_norm": 1.848891121802347, + "learning_rate": 3.7416082813989552e-06, + "loss": 1.1439, + "step": 175 + }, + { + "epoch": 0.02116274875248001, + "grad_norm": 2.3387683059081983, + "learning_rate": 3.745736189718439e-06, + "loss": 1.1331, + "step": 176 + }, + { + "epoch": 0.0212829916431191, + "grad_norm": 4.075370940193162, + "learning_rate": 3.749840710329894e-06, + "loss": 0.9657, + "step": 177 + }, + { + "epoch": 0.02140323453375819, + "grad_norm": 3.2070313986849084, + "learning_rate": 3.7539221067588938e-06, + "loss": 1.2199, + "step": 178 + }, + { + "epoch": 0.021523477424397284, + "grad_norm": 4.047065170709718, + "learning_rate": 3.757980638101964e-06, + "loss": 1.1725, + "step": 179 + }, + { + "epoch": 0.021643720315036374, + "grad_norm": 8.815023290957159, + "learning_rate": 3.7620165591252806e-06, + "loss": 1.1344, + "step": 180 + }, + { + "epoch": 0.021763963205675464, + "grad_norm": 2.0233312159133883, + "learning_rate": 3.766030120360636e-06, + "loss": 1.1813, + "step": 181 + }, + { + "epoch": 0.021884206096314557, + "grad_norm": 2.565662479475359, + "learning_rate": 3.7700215681987578e-06, + "loss": 1.1415, + "step": 182 + }, + { + "epoch": 0.022004448986953647, + "grad_norm": 1.9935154696195123, + "learning_rate": 3.7739911449800767e-06, + "loss": 1.0619, + "step": 183 + }, + { + "epoch": 0.022124691877592736, + "grad_norm": 2.178291995243183, + "learning_rate": 3.7779390890830114e-06, + "loss": 1.0461, + "step": 184 + }, + { + "epoch": 0.02224493476823183, + "grad_norm": 2.6571966840416397, + "learning_rate": 3.7818656350098723e-06, + "loss": 1.0977, + "step": 185 + }, + { + "epoch": 0.02236517765887092, + "grad_norm": 2.8710533917762966, + "learning_rate": 3.7857710134704447e-06, + "loss": 1.0117, + "step": 186 + }, + { + "epoch": 0.02248542054951001, + "grad_norm": 2.2656459412757948, + "learning_rate": 3.7896554514633234e-06, + "loss": 1.033, + "step": 187 + }, + { + "epoch": 0.022605663440149103, + "grad_norm": 1.8982643707828653, + "learning_rate": 3.7935191723550955e-06, + "loss": 1.0823, + "step": 188 + }, + { + "epoch": 0.022725906330788193, + "grad_norm": 2.218222112048049, + "learning_rate": 3.797362395957408e-06, + "loss": 1.1242, + "step": 189 + }, + { + "epoch": 0.022846149221427282, + "grad_norm": 2.3762079479784872, + "learning_rate": 3.8011853386020055e-06, + "loss": 1.0211, + "step": 190 + }, + { + "epoch": 0.022966392112066376, + "grad_norm": 3.246996445353412, + "learning_rate": 3.804988213213804e-06, + "loss": 1.1316, + "step": 191 + }, + { + "epoch": 0.023086635002705466, + "grad_norm": 1.0120673444376733, + "learning_rate": 3.808771229382049e-06, + "loss": 0.8696, + "step": 192 + }, + { + "epoch": 0.023206877893344555, + "grad_norm": 2.285916924009199, + "learning_rate": 3.8125345934296324e-06, + "loss": 1.082, + "step": 193 + }, + { + "epoch": 0.02332712078398365, + "grad_norm": 2.4932197179084676, + "learning_rate": 3.81627850848061e-06, + "loss": 1.1157, + "step": 194 + }, + { + "epoch": 0.02344736367462274, + "grad_norm": 2.487182182826841, + "learning_rate": 3.820003174525994e-06, + "loss": 1.1016, + "step": 195 + }, + { + "epoch": 0.02356760656526183, + "grad_norm": 2.707425263589443, + "learning_rate": 3.823708788487851e-06, + "loss": 1.0664, + "step": 196 + }, + { + "epoch": 0.02368784945590092, + "grad_norm": 1.7891304636210668, + "learning_rate": 3.827395544281781e-06, + "loss": 1.0852, + "step": 197 + }, + { + "epoch": 0.02380809234654001, + "grad_norm": 3.117158588476122, + "learning_rate": 3.831063632877802e-06, + "loss": 1.0276, + "step": 198 + }, + { + "epoch": 0.0239283352371791, + "grad_norm": 2.9301454720171516, + "learning_rate": 3.834713242359712e-06, + "loss": 0.9964, + "step": 199 + }, + { + "epoch": 0.02404857812781819, + "grad_norm": 3.882023997999616, + "learning_rate": 3.838344557982959e-06, + "loss": 1.1127, + "step": 200 + }, + { + "epoch": 0.024168821018457284, + "grad_norm": 2.782696979744489, + "learning_rate": 3.841957762231063e-06, + "loss": 1.0903, + "step": 201 + }, + { + "epoch": 0.024289063909096374, + "grad_norm": 2.251210099226226, + "learning_rate": 3.8455530348706454e-06, + "loss": 1.1141, + "step": 202 + }, + { + "epoch": 0.024409306799735464, + "grad_norm": 2.8447287415778644, + "learning_rate": 3.849130553005099e-06, + "loss": 1.0126, + "step": 203 + }, + { + "epoch": 0.024529549690374557, + "grad_norm": 2.053891916322952, + "learning_rate": 3.852690491126933e-06, + "loss": 1.0754, + "step": 204 + }, + { + "epoch": 0.024649792581013647, + "grad_norm": 3.294037379661311, + "learning_rate": 3.856233021168845e-06, + "loss": 1.1471, + "step": 205 + }, + { + "epoch": 0.024770035471652737, + "grad_norm": 2.1711689192369614, + "learning_rate": 3.859758312553544e-06, + "loss": 1.1541, + "step": 206 + }, + { + "epoch": 0.02489027836229183, + "grad_norm": 2.104755004615874, + "learning_rate": 3.8632665322423735e-06, + "loss": 1.1569, + "step": 207 + }, + { + "epoch": 0.02501052125293092, + "grad_norm": 1.7338187210556097, + "learning_rate": 3.866757844782762e-06, + "loss": 1.0967, + "step": 208 + }, + { + "epoch": 0.02513076414357001, + "grad_norm": 2.3507706332260154, + "learning_rate": 3.870232412354527e-06, + "loss": 1.1547, + "step": 209 + }, + { + "epoch": 0.025251007034209103, + "grad_norm": 2.1266577369279243, + "learning_rate": 3.873690394815086e-06, + "loss": 1.1642, + "step": 210 + }, + { + "epoch": 0.025371249924848193, + "grad_norm": 3.1221990621423514, + "learning_rate": 3.877131949743587e-06, + "loss": 1.1504, + "step": 211 + }, + { + "epoch": 0.025491492815487283, + "grad_norm": 2.3810793717896015, + "learning_rate": 3.880557232483993e-06, + "loss": 1.0215, + "step": 212 + }, + { + "epoch": 0.025611735706126376, + "grad_norm": 2.0422527743519274, + "learning_rate": 3.883966396187164e-06, + "loss": 1.1076, + "step": 213 + }, + { + "epoch": 0.025731978596765466, + "grad_norm": 1.971943702045024, + "learning_rate": 3.887359591851937e-06, + "loss": 1.1359, + "step": 214 + }, + { + "epoch": 0.025852221487404556, + "grad_norm": 1.6249173216694297, + "learning_rate": 3.890736968365265e-06, + "loss": 1.1605, + "step": 215 + }, + { + "epoch": 0.02597246437804365, + "grad_norm": 2.220820360939066, + "learning_rate": 3.894098672541412e-06, + "loss": 1.0877, + "step": 216 + }, + { + "epoch": 0.02609270726868274, + "grad_norm": 1.7777552537021033, + "learning_rate": 3.89744484916025e-06, + "loss": 0.992, + "step": 217 + }, + { + "epoch": 0.02621295015932183, + "grad_norm": 1.9118684611063708, + "learning_rate": 3.900775641004673e-06, + "loss": 1.1155, + "step": 218 + }, + { + "epoch": 0.026333193049960922, + "grad_norm": 2.9243597389535543, + "learning_rate": 3.904091188897156e-06, + "loss": 0.9774, + "step": 219 + }, + { + "epoch": 0.026453435940600012, + "grad_norm": 2.120914121436852, + "learning_rate": 3.90739163173548e-06, + "loss": 1.058, + "step": 220 + }, + { + "epoch": 0.026573678831239102, + "grad_norm": 3.948500510385883, + "learning_rate": 3.910677106527646e-06, + "loss": 1.1176, + "step": 221 + }, + { + "epoch": 0.026693921721878195, + "grad_norm": 2.191862932514075, + "learning_rate": 3.913947748426004e-06, + "loss": 1.0818, + "step": 222 + }, + { + "epoch": 0.026814164612517285, + "grad_norm": 3.5001781963217042, + "learning_rate": 3.9172036907606136e-06, + "loss": 1.0023, + "step": 223 + }, + { + "epoch": 0.026934407503156375, + "grad_norm": 2.011968919328161, + "learning_rate": 3.920445065071855e-06, + "loss": 1.1895, + "step": 224 + }, + { + "epoch": 0.027054650393795468, + "grad_norm": 2.32359354281455, + "learning_rate": 3.923672001142322e-06, + "loss": 1.0387, + "step": 225 + }, + { + "epoch": 0.027174893284434558, + "grad_norm": 2.0422145990559883, + "learning_rate": 3.926884627027996e-06, + "loss": 1.083, + "step": 226 + }, + { + "epoch": 0.027295136175073648, + "grad_norm": 1.9495443251864522, + "learning_rate": 3.930083069088744e-06, + "loss": 1.0146, + "step": 227 + }, + { + "epoch": 0.02741537906571274, + "grad_norm": 0.985590441720227, + "learning_rate": 3.933267452018137e-06, + "loss": 0.8307, + "step": 228 + }, + { + "epoch": 0.02753562195635183, + "grad_norm": 3.081332941080308, + "learning_rate": 3.936437898872622e-06, + "loss": 1.0791, + "step": 229 + }, + { + "epoch": 0.02765586484699092, + "grad_norm": 6.68596424758029, + "learning_rate": 3.9395945311000525e-06, + "loss": 1.0321, + "step": 230 + }, + { + "epoch": 0.027776107737630014, + "grad_norm": 2.3299247390641975, + "learning_rate": 3.942737468567608e-06, + "loss": 1.1473, + "step": 231 + }, + { + "epoch": 0.027896350628269104, + "grad_norm": 2.019771480475825, + "learning_rate": 3.9458668295891026e-06, + "loss": 1.0994, + "step": 232 + }, + { + "epoch": 0.028016593518908194, + "grad_norm": 2.281991185009954, + "learning_rate": 3.948982730951712e-06, + "loss": 1.1042, + "step": 233 + }, + { + "epoch": 0.028136836409547287, + "grad_norm": 2.255899641990456, + "learning_rate": 3.9520852879421254e-06, + "loss": 1.061, + "step": 234 + }, + { + "epoch": 0.028257079300186377, + "grad_norm": 2.258567921277373, + "learning_rate": 3.955174614372137e-06, + "loss": 1.0518, + "step": 235 + }, + { + "epoch": 0.028377322190825467, + "grad_norm": 2.3995737372783497, + "learning_rate": 3.9582508226037045e-06, + "loss": 1.0829, + "step": 236 + }, + { + "epoch": 0.02849756508146456, + "grad_norm": 2.6421134410474827, + "learning_rate": 3.9613140235734636e-06, + "loss": 1.1789, + "step": 237 + }, + { + "epoch": 0.02861780797210365, + "grad_norm": 3.0319367250266667, + "learning_rate": 3.96436432681674e-06, + "loss": 1.0487, + "step": 238 + }, + { + "epoch": 0.02873805086274274, + "grad_norm": 1.9760458048480722, + "learning_rate": 3.967401840491044e-06, + "loss": 1.1309, + "step": 239 + }, + { + "epoch": 0.028858293753381833, + "grad_norm": 2.706768777296117, + "learning_rate": 3.97042667139909e-06, + "loss": 1.1157, + "step": 240 + }, + { + "epoch": 0.028978536644020923, + "grad_norm": 2.414288359017427, + "learning_rate": 3.973438925011327e-06, + "loss": 1.1137, + "step": 241 + }, + { + "epoch": 0.029098779534660012, + "grad_norm": 3.661478591508384, + "learning_rate": 3.976438705488002e-06, + "loss": 1.1589, + "step": 242 + }, + { + "epoch": 0.029219022425299106, + "grad_norm": 2.932954558568365, + "learning_rate": 3.9794261157007744e-06, + "loss": 1.1712, + "step": 243 + }, + { + "epoch": 0.029339265315938196, + "grad_norm": 2.4307327230157876, + "learning_rate": 3.982401257253887e-06, + "loss": 1.0882, + "step": 244 + }, + { + "epoch": 0.029459508206577285, + "grad_norm": 16.05586408401538, + "learning_rate": 3.985364230504893e-06, + "loss": 1.139, + "step": 245 + }, + { + "epoch": 0.02957975109721638, + "grad_norm": 2.081016192892445, + "learning_rate": 3.988315134584976e-06, + "loss": 1.0818, + "step": 246 + }, + { + "epoch": 0.02969999398785547, + "grad_norm": 1.8376220766374707, + "learning_rate": 3.991254067418851e-06, + "loss": 1.044, + "step": 247 + }, + { + "epoch": 0.02982023687849456, + "grad_norm": 16.84156029221081, + "learning_rate": 3.994181125744254e-06, + "loss": 1.069, + "step": 248 + }, + { + "epoch": 0.02994047976913365, + "grad_norm": 2.001001345777464, + "learning_rate": 3.99709640513106e-06, + "loss": 0.9802, + "step": 249 + }, + { + "epoch": 0.03006072265977274, + "grad_norm": 2.226674716196339, + "learning_rate": 4e-06, + "loss": 1.0952, + "step": 250 + }, + { + "epoch": 0.03018096555041183, + "grad_norm": 3.088849588947654, + "learning_rate": 3.999999848300794e-06, + "loss": 1.1231, + "step": 251 + }, + { + "epoch": 0.030301208441050925, + "grad_norm": 1.8019422648640329, + "learning_rate": 3.999999393203203e-06, + "loss": 1.1285, + "step": 252 + }, + { + "epoch": 0.030421451331690014, + "grad_norm": 2.016507096190846, + "learning_rate": 3.999998634707293e-06, + "loss": 1.0912, + "step": 253 + }, + { + "epoch": 0.030541694222329104, + "grad_norm": 2.371151667332182, + "learning_rate": 3.999997572813182e-06, + "loss": 1.2023, + "step": 254 + }, + { + "epoch": 0.030661937112968194, + "grad_norm": 2.196394792711278, + "learning_rate": 3.999996207521028e-06, + "loss": 1.1153, + "step": 255 + }, + { + "epoch": 0.030782180003607287, + "grad_norm": 2.4331730315129128, + "learning_rate": 3.999994538831039e-06, + "loss": 1.0597, + "step": 256 + }, + { + "epoch": 0.030902422894246377, + "grad_norm": 2.6215321476937112, + "learning_rate": 3.99999256674347e-06, + "loss": 1.0973, + "step": 257 + }, + { + "epoch": 0.031022665784885467, + "grad_norm": 1.0149173912481964, + "learning_rate": 3.999990291258618e-06, + "loss": 0.773, + "step": 258 + }, + { + "epoch": 0.03114290867552456, + "grad_norm": 3.0336476704582163, + "learning_rate": 3.999987712376829e-06, + "loss": 1.1059, + "step": 259 + }, + { + "epoch": 0.031263151566163654, + "grad_norm": 2.3187303896997573, + "learning_rate": 3.999984830098494e-06, + "loss": 1.0613, + "step": 260 + }, + { + "epoch": 0.03138339445680274, + "grad_norm": 4.480838244899399, + "learning_rate": 3.999981644424051e-06, + "loss": 1.2199, + "step": 261 + }, + { + "epoch": 0.03150363734744183, + "grad_norm": 3.178269728644436, + "learning_rate": 3.999978155353982e-06, + "loss": 1.1002, + "step": 262 + }, + { + "epoch": 0.03162388023808092, + "grad_norm": 3.084315547350803, + "learning_rate": 3.9999743628888186e-06, + "loss": 1.0423, + "step": 263 + }, + { + "epoch": 0.03174412312872001, + "grad_norm": 2.7010140528053808, + "learning_rate": 3.999970267029133e-06, + "loss": 1.1329, + "step": 264 + }, + { + "epoch": 0.0318643660193591, + "grad_norm": 1.8016371937310471, + "learning_rate": 3.999965867775548e-06, + "loss": 1.0369, + "step": 265 + }, + { + "epoch": 0.0319846089099982, + "grad_norm": 5.223552572732713, + "learning_rate": 3.9999611651287315e-06, + "loss": 1.1061, + "step": 266 + }, + { + "epoch": 0.03210485180063729, + "grad_norm": 2.3825186126078024, + "learning_rate": 3.999956159089396e-06, + "loss": 1.0243, + "step": 267 + }, + { + "epoch": 0.03222509469127638, + "grad_norm": 3.189677523155153, + "learning_rate": 3.999950849658302e-06, + "loss": 1.0329, + "step": 268 + }, + { + "epoch": 0.03234533758191547, + "grad_norm": 2.0739350493072624, + "learning_rate": 3.999945236836254e-06, + "loss": 1.0804, + "step": 269 + }, + { + "epoch": 0.03246558047255456, + "grad_norm": 3.0100362112712302, + "learning_rate": 3.999939320624103e-06, + "loss": 1.1827, + "step": 270 + }, + { + "epoch": 0.03258582336319365, + "grad_norm": 1.919198009363679, + "learning_rate": 3.999933101022749e-06, + "loss": 1.1384, + "step": 271 + }, + { + "epoch": 0.032706066253832745, + "grad_norm": 1.8585195416728297, + "learning_rate": 3.999926578033132e-06, + "loss": 1.1065, + "step": 272 + }, + { + "epoch": 0.032826309144471835, + "grad_norm": 2.0635786992588065, + "learning_rate": 3.999919751656244e-06, + "loss": 0.868, + "step": 273 + }, + { + "epoch": 0.032946552035110925, + "grad_norm": 2.1713708525709126, + "learning_rate": 3.9999126218931195e-06, + "loss": 0.9941, + "step": 274 + }, + { + "epoch": 0.033066794925750015, + "grad_norm": 2.41989133259832, + "learning_rate": 3.99990518874484e-06, + "loss": 1.1335, + "step": 275 + }, + { + "epoch": 0.033187037816389105, + "grad_norm": 2.490992914496796, + "learning_rate": 3.999897452212534e-06, + "loss": 1.1595, + "step": 276 + }, + { + "epoch": 0.033307280707028195, + "grad_norm": 2.4730688055256396, + "learning_rate": 3.999889412297374e-06, + "loss": 1.2397, + "step": 277 + }, + { + "epoch": 0.03342752359766729, + "grad_norm": 2.6262383037962143, + "learning_rate": 3.999881069000581e-06, + "loss": 1.0294, + "step": 278 + }, + { + "epoch": 0.03354776648830638, + "grad_norm": 3.3173987674862864, + "learning_rate": 3.99987242232342e-06, + "loss": 1.1084, + "step": 279 + }, + { + "epoch": 0.03366800937894547, + "grad_norm": 2.0775085911762514, + "learning_rate": 3.9998634722672026e-06, + "loss": 1.0363, + "step": 280 + }, + { + "epoch": 0.03378825226958456, + "grad_norm": 2.3524356881086064, + "learning_rate": 3.999854218833286e-06, + "loss": 1.0224, + "step": 281 + }, + { + "epoch": 0.03390849516022365, + "grad_norm": 1.9030584915641593, + "learning_rate": 3.999844662023075e-06, + "loss": 1.0576, + "step": 282 + }, + { + "epoch": 0.03402873805086274, + "grad_norm": 1.8722594259674004, + "learning_rate": 3.999834801838018e-06, + "loss": 1.1623, + "step": 283 + }, + { + "epoch": 0.03414898094150183, + "grad_norm": 2.588430810367276, + "learning_rate": 3.9998246382796115e-06, + "loss": 0.9776, + "step": 284 + }, + { + "epoch": 0.03426922383214093, + "grad_norm": 2.649787131342943, + "learning_rate": 3.999814171349399e-06, + "loss": 1.1469, + "step": 285 + }, + { + "epoch": 0.03438946672278002, + "grad_norm": 2.23999958753582, + "learning_rate": 3.9998034010489655e-06, + "loss": 0.9753, + "step": 286 + }, + { + "epoch": 0.03450970961341911, + "grad_norm": 2.603101578660907, + "learning_rate": 3.999792327379946e-06, + "loss": 0.9958, + "step": 287 + }, + { + "epoch": 0.034629952504058197, + "grad_norm": 2.5214815383528775, + "learning_rate": 3.999780950344021e-06, + "loss": 1.2003, + "step": 288 + }, + { + "epoch": 0.034750195394697286, + "grad_norm": 4.100512095779238, + "learning_rate": 3.999769269942916e-06, + "loss": 1.0667, + "step": 289 + }, + { + "epoch": 0.034870438285336376, + "grad_norm": 1.7739040752482569, + "learning_rate": 3.999757286178402e-06, + "loss": 1.051, + "step": 290 + }, + { + "epoch": 0.03499068117597547, + "grad_norm": 2.4901326955344425, + "learning_rate": 3.999744999052299e-06, + "loss": 1.1468, + "step": 291 + }, + { + "epoch": 0.03511092406661456, + "grad_norm": 0.9541985141375584, + "learning_rate": 3.9997324085664675e-06, + "loss": 0.85, + "step": 292 + }, + { + "epoch": 0.03523116695725365, + "grad_norm": 7.593713830932382, + "learning_rate": 3.999719514722821e-06, + "loss": 1.1556, + "step": 293 + }, + { + "epoch": 0.03535140984789274, + "grad_norm": 2.3159253295035094, + "learning_rate": 3.999706317523314e-06, + "loss": 0.9885, + "step": 294 + }, + { + "epoch": 0.03547165273853183, + "grad_norm": 2.1625269320359934, + "learning_rate": 3.999692816969948e-06, + "loss": 1.1013, + "step": 295 + }, + { + "epoch": 0.03559189562917092, + "grad_norm": 1.0264952313869362, + "learning_rate": 3.999679013064772e-06, + "loss": 0.9319, + "step": 296 + }, + { + "epoch": 0.03571213851981002, + "grad_norm": 3.9604082525178295, + "learning_rate": 3.99966490580988e-06, + "loss": 1.0974, + "step": 297 + }, + { + "epoch": 0.03583238141044911, + "grad_norm": 2.6641357053839543, + "learning_rate": 3.999650495207411e-06, + "loss": 0.894, + "step": 298 + }, + { + "epoch": 0.0359526243010882, + "grad_norm": 3.2114572220432382, + "learning_rate": 3.999635781259553e-06, + "loss": 1.144, + "step": 299 + }, + { + "epoch": 0.03607286719172729, + "grad_norm": 0.9191925008249607, + "learning_rate": 3.999620763968535e-06, + "loss": 0.7613, + "step": 300 + }, + { + "epoch": 0.03619311008236638, + "grad_norm": 1.9097006162712054, + "learning_rate": 3.999605443336638e-06, + "loss": 1.1061, + "step": 301 + }, + { + "epoch": 0.03631335297300547, + "grad_norm": 2.3184044181992243, + "learning_rate": 3.999589819366185e-06, + "loss": 1.1293, + "step": 302 + }, + { + "epoch": 0.036433595863644565, + "grad_norm": 2.0373710544715786, + "learning_rate": 3.999573892059547e-06, + "loss": 1.0846, + "step": 303 + }, + { + "epoch": 0.036553838754283655, + "grad_norm": 2.4526512514656296, + "learning_rate": 3.999557661419138e-06, + "loss": 1.0513, + "step": 304 + }, + { + "epoch": 0.036674081644922744, + "grad_norm": 1.9077485244859247, + "learning_rate": 3.9995411274474225e-06, + "loss": 1.051, + "step": 305 + }, + { + "epoch": 0.036794324535561834, + "grad_norm": 2.7330258774158644, + "learning_rate": 3.999524290146908e-06, + "loss": 1.0532, + "step": 306 + }, + { + "epoch": 0.036914567426200924, + "grad_norm": 8.034916497243502, + "learning_rate": 3.9995071495201485e-06, + "loss": 1.1661, + "step": 307 + }, + { + "epoch": 0.037034810316840014, + "grad_norm": 3.0204047238813745, + "learning_rate": 3.999489705569744e-06, + "loss": 1.2185, + "step": 308 + }, + { + "epoch": 0.03715505320747911, + "grad_norm": 2.4293216540282327, + "learning_rate": 3.999471958298341e-06, + "loss": 1.1034, + "step": 309 + }, + { + "epoch": 0.0372752960981182, + "grad_norm": 2.2853491546850297, + "learning_rate": 3.999453907708631e-06, + "loss": 0.9978, + "step": 310 + }, + { + "epoch": 0.03739553898875729, + "grad_norm": 2.1730830722875045, + "learning_rate": 3.999435553803353e-06, + "loss": 1.0554, + "step": 311 + }, + { + "epoch": 0.03751578187939638, + "grad_norm": 2.8471505091456817, + "learning_rate": 3.999416896585292e-06, + "loss": 1.0757, + "step": 312 + }, + { + "epoch": 0.03763602477003547, + "grad_norm": 3.001833832057605, + "learning_rate": 3.9993979360572775e-06, + "loss": 1.0955, + "step": 313 + }, + { + "epoch": 0.03775626766067456, + "grad_norm": 2.8504755326127476, + "learning_rate": 3.999378672222185e-06, + "loss": 1.0685, + "step": 314 + }, + { + "epoch": 0.03787651055131366, + "grad_norm": 2.240501274832133, + "learning_rate": 3.9993591050829385e-06, + "loss": 1.0673, + "step": 315 + }, + { + "epoch": 0.037996753441952746, + "grad_norm": 2.413914845713232, + "learning_rate": 3.999339234642506e-06, + "loss": 1.033, + "step": 316 + }, + { + "epoch": 0.038116996332591836, + "grad_norm": 2.1898154946558317, + "learning_rate": 3.9993190609038994e-06, + "loss": 1.0739, + "step": 317 + }, + { + "epoch": 0.038237239223230926, + "grad_norm": 2.4002790805361007, + "learning_rate": 3.999298583870182e-06, + "loss": 1.069, + "step": 318 + }, + { + "epoch": 0.038357482113870016, + "grad_norm": 1.8663035578877103, + "learning_rate": 3.999277803544458e-06, + "loss": 1.011, + "step": 319 + }, + { + "epoch": 0.038477725004509106, + "grad_norm": 1.079477650278959, + "learning_rate": 3.999256719929882e-06, + "loss": 0.8631, + "step": 320 + }, + { + "epoch": 0.0385979678951482, + "grad_norm": 1.2531875499610992, + "learning_rate": 3.999235333029651e-06, + "loss": 0.9093, + "step": 321 + }, + { + "epoch": 0.03871821078578729, + "grad_norm": 2.1610739002222203, + "learning_rate": 3.999213642847009e-06, + "loss": 1.0585, + "step": 322 + }, + { + "epoch": 0.03883845367642638, + "grad_norm": 2.4278805014349545, + "learning_rate": 3.999191649385247e-06, + "loss": 1.1522, + "step": 323 + }, + { + "epoch": 0.03895869656706547, + "grad_norm": 0.9169684946332638, + "learning_rate": 3.999169352647702e-06, + "loss": 0.8364, + "step": 324 + }, + { + "epoch": 0.03907893945770456, + "grad_norm": 2.0425279189107868, + "learning_rate": 3.999146752637755e-06, + "loss": 1.0679, + "step": 325 + }, + { + "epoch": 0.03919918234834365, + "grad_norm": 2.4476908369935524, + "learning_rate": 3.999123849358836e-06, + "loss": 1.1383, + "step": 326 + }, + { + "epoch": 0.03931942523898275, + "grad_norm": 2.172898634497146, + "learning_rate": 3.999100642814418e-06, + "loss": 0.9872, + "step": 327 + }, + { + "epoch": 0.03943966812962184, + "grad_norm": 2.2509780453492945, + "learning_rate": 3.999077133008022e-06, + "loss": 1.1242, + "step": 328 + }, + { + "epoch": 0.03955991102026093, + "grad_norm": 2.259762670833203, + "learning_rate": 3.9990533199432145e-06, + "loss": 1.1458, + "step": 329 + }, + { + "epoch": 0.03968015391090002, + "grad_norm": 2.5177678997762194, + "learning_rate": 3.999029203623608e-06, + "loss": 0.9919, + "step": 330 + }, + { + "epoch": 0.03980039680153911, + "grad_norm": 2.6593822187780716, + "learning_rate": 3.99900478405286e-06, + "loss": 1.1066, + "step": 331 + }, + { + "epoch": 0.0399206396921782, + "grad_norm": 4.16731063952624, + "learning_rate": 3.998980061234676e-06, + "loss": 1.0632, + "step": 332 + }, + { + "epoch": 0.040040882582817294, + "grad_norm": 13.776232301177838, + "learning_rate": 3.9989550351728055e-06, + "loss": 0.9999, + "step": 333 + }, + { + "epoch": 0.040161125473456384, + "grad_norm": 2.6383520874237263, + "learning_rate": 3.998929705871046e-06, + "loss": 1.0809, + "step": 334 + }, + { + "epoch": 0.040281368364095474, + "grad_norm": 3.0605543534528925, + "learning_rate": 3.99890407333324e-06, + "loss": 1.128, + "step": 335 + }, + { + "epoch": 0.040401611254734564, + "grad_norm": 5.012484984720244, + "learning_rate": 3.998878137563275e-06, + "loss": 1.1107, + "step": 336 + }, + { + "epoch": 0.040521854145373654, + "grad_norm": 3.156787335574485, + "learning_rate": 3.998851898565085e-06, + "loss": 1.0874, + "step": 337 + }, + { + "epoch": 0.04064209703601274, + "grad_norm": 2.2050659071226506, + "learning_rate": 3.998825356342653e-06, + "loss": 1.07, + "step": 338 + }, + { + "epoch": 0.04076233992665183, + "grad_norm": 2.679408467615217, + "learning_rate": 3.998798510900003e-06, + "loss": 0.97, + "step": 339 + }, + { + "epoch": 0.04088258281729093, + "grad_norm": 2.203940058154428, + "learning_rate": 3.998771362241207e-06, + "loss": 1.0781, + "step": 340 + }, + { + "epoch": 0.04100282570793002, + "grad_norm": 2.805032012947169, + "learning_rate": 3.998743910370385e-06, + "loss": 1.1168, + "step": 341 + }, + { + "epoch": 0.04112306859856911, + "grad_norm": 3.4395844122000994, + "learning_rate": 3.998716155291702e-06, + "loss": 0.9721, + "step": 342 + }, + { + "epoch": 0.0412433114892082, + "grad_norm": 3.113317089759988, + "learning_rate": 3.998688097009366e-06, + "loss": 1.1435, + "step": 343 + }, + { + "epoch": 0.04136355437984729, + "grad_norm": 2.3853083312464083, + "learning_rate": 3.998659735527636e-06, + "loss": 1.04, + "step": 344 + }, + { + "epoch": 0.04148379727048638, + "grad_norm": 1.997950576743728, + "learning_rate": 3.998631070850813e-06, + "loss": 1.0165, + "step": 345 + }, + { + "epoch": 0.041604040161125476, + "grad_norm": 2.5485207324817196, + "learning_rate": 3.9986021029832455e-06, + "loss": 1.0681, + "step": 346 + }, + { + "epoch": 0.041724283051764566, + "grad_norm": 3.065114209534887, + "learning_rate": 3.9985728319293285e-06, + "loss": 1.1543, + "step": 347 + }, + { + "epoch": 0.041844525942403656, + "grad_norm": 2.4085417893775434, + "learning_rate": 3.998543257693501e-06, + "loss": 1.0903, + "step": 348 + }, + { + "epoch": 0.041964768833042745, + "grad_norm": 2.4785737255552225, + "learning_rate": 3.998513380280251e-06, + "loss": 1.1187, + "step": 349 + }, + { + "epoch": 0.042085011723681835, + "grad_norm": 2.592268739132065, + "learning_rate": 3.99848319969411e-06, + "loss": 1.1847, + "step": 350 + }, + { + "epoch": 0.042205254614320925, + "grad_norm": 2.105387556955757, + "learning_rate": 3.9984527159396564e-06, + "loss": 1.0267, + "step": 351 + }, + { + "epoch": 0.04232549750496002, + "grad_norm": 2.6977390147354834, + "learning_rate": 3.9984219290215154e-06, + "loss": 1.0823, + "step": 352 + }, + { + "epoch": 0.04244574039559911, + "grad_norm": 1.7761639714196267, + "learning_rate": 3.998390838944356e-06, + "loss": 1.1294, + "step": 353 + }, + { + "epoch": 0.0425659832862382, + "grad_norm": 2.7012732894295963, + "learning_rate": 3.998359445712895e-06, + "loss": 1.1418, + "step": 354 + }, + { + "epoch": 0.04268622617687729, + "grad_norm": 2.953326183106641, + "learning_rate": 3.9983277493318955e-06, + "loss": 1.0498, + "step": 355 + }, + { + "epoch": 0.04280646906751638, + "grad_norm": 1.8602446758172029, + "learning_rate": 3.998295749806165e-06, + "loss": 1.0515, + "step": 356 + }, + { + "epoch": 0.04292671195815547, + "grad_norm": 2.2831766014418555, + "learning_rate": 3.998263447140558e-06, + "loss": 1.0743, + "step": 357 + }, + { + "epoch": 0.04304695484879457, + "grad_norm": 1.9108035623144684, + "learning_rate": 3.998230841339976e-06, + "loss": 1.0567, + "step": 358 + }, + { + "epoch": 0.04316719773943366, + "grad_norm": 2.3887122985962725, + "learning_rate": 3.998197932409363e-06, + "loss": 1.087, + "step": 359 + }, + { + "epoch": 0.04328744063007275, + "grad_norm": 3.133971496958771, + "learning_rate": 3.9981647203537125e-06, + "loss": 1.1025, + "step": 360 + }, + { + "epoch": 0.04340768352071184, + "grad_norm": 2.6767000945756454, + "learning_rate": 3.998131205178063e-06, + "loss": 1.1962, + "step": 361 + }, + { + "epoch": 0.04352792641135093, + "grad_norm": 4.991015563854973, + "learning_rate": 3.998097386887498e-06, + "loss": 1.0042, + "step": 362 + }, + { + "epoch": 0.04364816930199002, + "grad_norm": 2.143759889274008, + "learning_rate": 3.998063265487148e-06, + "loss": 1.0905, + "step": 363 + }, + { + "epoch": 0.043768412192629114, + "grad_norm": 5.07186475601149, + "learning_rate": 3.99802884098219e-06, + "loss": 1.0497, + "step": 364 + }, + { + "epoch": 0.043888655083268203, + "grad_norm": 2.6912012561967527, + "learning_rate": 3.997994113377845e-06, + "loss": 1.0629, + "step": 365 + }, + { + "epoch": 0.04400889797390729, + "grad_norm": 5.447277179868526, + "learning_rate": 3.9979590826793815e-06, + "loss": 1.0709, + "step": 366 + }, + { + "epoch": 0.04412914086454638, + "grad_norm": 2.272270715978965, + "learning_rate": 3.997923748892113e-06, + "loss": 1.0515, + "step": 367 + }, + { + "epoch": 0.04424938375518547, + "grad_norm": 1.7703779944133478, + "learning_rate": 3.9978881120214015e-06, + "loss": 1.126, + "step": 368 + }, + { + "epoch": 0.04436962664582456, + "grad_norm": 2.117902909356204, + "learning_rate": 3.997852172072652e-06, + "loss": 1.0311, + "step": 369 + }, + { + "epoch": 0.04448986953646366, + "grad_norm": 3.051202762024004, + "learning_rate": 3.9978159290513155e-06, + "loss": 1.129, + "step": 370 + }, + { + "epoch": 0.04461011242710275, + "grad_norm": 1.7613548021290755, + "learning_rate": 3.997779382962892e-06, + "loss": 1.0406, + "step": 371 + }, + { + "epoch": 0.04473035531774184, + "grad_norm": 9.932418579720322, + "learning_rate": 3.997742533812924e-06, + "loss": 0.9773, + "step": 372 + }, + { + "epoch": 0.04485059820838093, + "grad_norm": 2.816498944030921, + "learning_rate": 3.997705381607001e-06, + "loss": 1.1622, + "step": 373 + }, + { + "epoch": 0.04497084109902002, + "grad_norm": 1.0360534810176885, + "learning_rate": 3.997667926350761e-06, + "loss": 0.8418, + "step": 374 + }, + { + "epoch": 0.04509108398965911, + "grad_norm": 0.8934686671475094, + "learning_rate": 3.997630168049886e-06, + "loss": 0.816, + "step": 375 + }, + { + "epoch": 0.045211326880298205, + "grad_norm": 1.8771816277320832, + "learning_rate": 3.997592106710101e-06, + "loss": 1.0136, + "step": 376 + }, + { + "epoch": 0.045331569770937295, + "grad_norm": 4.5462962827720625, + "learning_rate": 3.997553742337182e-06, + "loss": 0.8981, + "step": 377 + }, + { + "epoch": 0.045451812661576385, + "grad_norm": 2.074965060755707, + "learning_rate": 3.997515074936949e-06, + "loss": 1.154, + "step": 378 + }, + { + "epoch": 0.045572055552215475, + "grad_norm": 2.873564983981237, + "learning_rate": 3.997476104515268e-06, + "loss": 1.1035, + "step": 379 + }, + { + "epoch": 0.045692298442854565, + "grad_norm": 2.2880570739149912, + "learning_rate": 3.9974368310780485e-06, + "loss": 1.0135, + "step": 380 + }, + { + "epoch": 0.045812541333493655, + "grad_norm": 6.9853498559387335, + "learning_rate": 3.997397254631251e-06, + "loss": 0.9841, + "step": 381 + }, + { + "epoch": 0.04593278422413275, + "grad_norm": 0.8238353108526326, + "learning_rate": 3.997357375180878e-06, + "loss": 0.8386, + "step": 382 + }, + { + "epoch": 0.04605302711477184, + "grad_norm": 6.903860930367124, + "learning_rate": 3.997317192732979e-06, + "loss": 0.9929, + "step": 383 + }, + { + "epoch": 0.04617327000541093, + "grad_norm": 1.9952764760277697, + "learning_rate": 3.99727670729365e-06, + "loss": 1.0606, + "step": 384 + }, + { + "epoch": 0.04629351289605002, + "grad_norm": 2.003718019683604, + "learning_rate": 3.997235918869033e-06, + "loss": 1.02, + "step": 385 + }, + { + "epoch": 0.04641375578668911, + "grad_norm": 5.436168288342413, + "learning_rate": 3.997194827465315e-06, + "loss": 1.0646, + "step": 386 + }, + { + "epoch": 0.0465339986773282, + "grad_norm": 2.900657769946681, + "learning_rate": 3.997153433088728e-06, + "loss": 1.154, + "step": 387 + }, + { + "epoch": 0.0466542415679673, + "grad_norm": 2.3419714728164567, + "learning_rate": 3.997111735745554e-06, + "loss": 1.0525, + "step": 388 + }, + { + "epoch": 0.04677448445860639, + "grad_norm": 2.0767076290901403, + "learning_rate": 3.997069735442118e-06, + "loss": 1.0647, + "step": 389 + }, + { + "epoch": 0.04689472734924548, + "grad_norm": 1.5888634361030545, + "learning_rate": 3.997027432184792e-06, + "loss": 1.0427, + "step": 390 + }, + { + "epoch": 0.04701497023988457, + "grad_norm": 2.0442484536535406, + "learning_rate": 3.99698482597999e-06, + "loss": 1.1312, + "step": 391 + }, + { + "epoch": 0.04713521313052366, + "grad_norm": 0.8662233921604967, + "learning_rate": 3.99694191683418e-06, + "loss": 0.877, + "step": 392 + }, + { + "epoch": 0.047255456021162746, + "grad_norm": 2.4108851982767696, + "learning_rate": 3.996898704753867e-06, + "loss": 1.0562, + "step": 393 + }, + { + "epoch": 0.04737569891180184, + "grad_norm": 2.743216384057361, + "learning_rate": 3.996855189745609e-06, + "loss": 1.1153, + "step": 394 + }, + { + "epoch": 0.04749594180244093, + "grad_norm": 2.082159843553137, + "learning_rate": 3.996811371816007e-06, + "loss": 1.1651, + "step": 395 + }, + { + "epoch": 0.04761618469308002, + "grad_norm": 2.191036091105497, + "learning_rate": 3.996767250971707e-06, + "loss": 1.0183, + "step": 396 + }, + { + "epoch": 0.04773642758371911, + "grad_norm": 4.632999615806533, + "learning_rate": 3.996722827219403e-06, + "loss": 1.1073, + "step": 397 + }, + { + "epoch": 0.0478566704743582, + "grad_norm": 2.674099982898242, + "learning_rate": 3.996678100565833e-06, + "loss": 1.0646, + "step": 398 + }, + { + "epoch": 0.04797691336499729, + "grad_norm": 2.727056857437837, + "learning_rate": 3.996633071017783e-06, + "loss": 1.1235, + "step": 399 + }, + { + "epoch": 0.04809715625563638, + "grad_norm": 8.592359617638692, + "learning_rate": 3.996587738582084e-06, + "loss": 1.053, + "step": 400 + }, + { + "epoch": 0.04821739914627548, + "grad_norm": 4.866173015449017, + "learning_rate": 3.9965421032656115e-06, + "loss": 1.1006, + "step": 401 + }, + { + "epoch": 0.04833764203691457, + "grad_norm": 2.425981055736872, + "learning_rate": 3.99649616507529e-06, + "loss": 1.1755, + "step": 402 + }, + { + "epoch": 0.04845788492755366, + "grad_norm": 0.8908005926136227, + "learning_rate": 3.996449924018088e-06, + "loss": 0.8696, + "step": 403 + }, + { + "epoch": 0.04857812781819275, + "grad_norm": 2.020642209273262, + "learning_rate": 3.99640338010102e-06, + "loss": 1.035, + "step": 404 + }, + { + "epoch": 0.04869837070883184, + "grad_norm": 2.8808448092852115, + "learning_rate": 3.996356533331146e-06, + "loss": 1.0268, + "step": 405 + }, + { + "epoch": 0.04881861359947093, + "grad_norm": 3.6119357186445216, + "learning_rate": 3.996309383715573e-06, + "loss": 0.8519, + "step": 406 + }, + { + "epoch": 0.048938856490110025, + "grad_norm": 2.448694200332733, + "learning_rate": 3.996261931261454e-06, + "loss": 0.9742, + "step": 407 + }, + { + "epoch": 0.049059099380749115, + "grad_norm": 2.0670053230220797, + "learning_rate": 3.996214175975987e-06, + "loss": 1.1022, + "step": 408 + }, + { + "epoch": 0.049179342271388204, + "grad_norm": 2.547480786132231, + "learning_rate": 3.996166117866417e-06, + "loss": 1.0293, + "step": 409 + }, + { + "epoch": 0.049299585162027294, + "grad_norm": 4.29148098935918, + "learning_rate": 3.996117756940035e-06, + "loss": 1.1055, + "step": 410 + }, + { + "epoch": 0.049419828052666384, + "grad_norm": 2.594599113736964, + "learning_rate": 3.996069093204175e-06, + "loss": 1.211, + "step": 411 + }, + { + "epoch": 0.049540070943305474, + "grad_norm": 2.475385223618752, + "learning_rate": 3.996020126666221e-06, + "loss": 1.121, + "step": 412 + }, + { + "epoch": 0.04966031383394457, + "grad_norm": 2.7552489558339857, + "learning_rate": 3.995970857333601e-06, + "loss": 1.0624, + "step": 413 + }, + { + "epoch": 0.04978055672458366, + "grad_norm": 2.8895915065467355, + "learning_rate": 3.995921285213789e-06, + "loss": 1.0343, + "step": 414 + }, + { + "epoch": 0.04990079961522275, + "grad_norm": 2.621882085597361, + "learning_rate": 3.995871410314305e-06, + "loss": 1.0425, + "step": 415 + }, + { + "epoch": 0.05002104250586184, + "grad_norm": 0.8976177954342806, + "learning_rate": 3.995821232642714e-06, + "loss": 0.8339, + "step": 416 + }, + { + "epoch": 0.05014128539650093, + "grad_norm": 4.724728062791972, + "learning_rate": 3.995770752206629e-06, + "loss": 1.0626, + "step": 417 + }, + { + "epoch": 0.05026152828714002, + "grad_norm": 8.267086945381669, + "learning_rate": 3.995719969013709e-06, + "loss": 1.2091, + "step": 418 + }, + { + "epoch": 0.05038177117777912, + "grad_norm": 2.9897436615134847, + "learning_rate": 3.995668883071655e-06, + "loss": 1.1018, + "step": 419 + }, + { + "epoch": 0.050502014068418206, + "grad_norm": 3.304040146498352, + "learning_rate": 3.995617494388219e-06, + "loss": 1.1492, + "step": 420 + }, + { + "epoch": 0.050622256959057296, + "grad_norm": 3.2105953284836835, + "learning_rate": 3.995565802971196e-06, + "loss": 1.0439, + "step": 421 + }, + { + "epoch": 0.050742499849696386, + "grad_norm": 2.415418719210259, + "learning_rate": 3.995513808828427e-06, + "loss": 0.9173, + "step": 422 + }, + { + "epoch": 0.050862742740335476, + "grad_norm": 2.6390470948116924, + "learning_rate": 3.9954615119678e-06, + "loss": 1.0049, + "step": 423 + }, + { + "epoch": 0.050982985630974566, + "grad_norm": 6.855096199367876, + "learning_rate": 3.995408912397248e-06, + "loss": 1.0437, + "step": 424 + }, + { + "epoch": 0.05110322852161366, + "grad_norm": 2.51404536154283, + "learning_rate": 3.99535601012475e-06, + "loss": 1.1721, + "step": 425 + }, + { + "epoch": 0.05122347141225275, + "grad_norm": 2.3083127233179632, + "learning_rate": 3.995302805158333e-06, + "loss": 0.9923, + "step": 426 + }, + { + "epoch": 0.05134371430289184, + "grad_norm": 2.1756248760360335, + "learning_rate": 3.9952492975060665e-06, + "loss": 1.0738, + "step": 427 + }, + { + "epoch": 0.05146395719353093, + "grad_norm": 2.7903618016706027, + "learning_rate": 3.995195487176067e-06, + "loss": 1.0921, + "step": 428 + }, + { + "epoch": 0.05158420008417002, + "grad_norm": 2.623820266836807, + "learning_rate": 3.995141374176499e-06, + "loss": 1.0939, + "step": 429 + }, + { + "epoch": 0.05170444297480911, + "grad_norm": 0.8788586073784298, + "learning_rate": 3.995086958515572e-06, + "loss": 0.8694, + "step": 430 + }, + { + "epoch": 0.05182468586544821, + "grad_norm": 0.8687530682187136, + "learning_rate": 3.995032240201538e-06, + "loss": 0.8373, + "step": 431 + }, + { + "epoch": 0.0519449287560873, + "grad_norm": 0.9377699145544351, + "learning_rate": 3.9949772192427e-06, + "loss": 0.8695, + "step": 432 + }, + { + "epoch": 0.05206517164672639, + "grad_norm": 2.241939411328094, + "learning_rate": 3.994921895647405e-06, + "loss": 1.0351, + "step": 433 + }, + { + "epoch": 0.05218541453736548, + "grad_norm": 0.8370574151813719, + "learning_rate": 3.994866269424043e-06, + "loss": 0.7914, + "step": 434 + }, + { + "epoch": 0.05230565742800457, + "grad_norm": 5.062139616018649, + "learning_rate": 3.9948103405810545e-06, + "loss": 1.019, + "step": 435 + }, + { + "epoch": 0.05242590031864366, + "grad_norm": 2.081687798719854, + "learning_rate": 3.994754109126923e-06, + "loss": 1.0994, + "step": 436 + }, + { + "epoch": 0.052546143209282754, + "grad_norm": 1.8203162328329818, + "learning_rate": 3.994697575070181e-06, + "loss": 1.1718, + "step": 437 + }, + { + "epoch": 0.052666386099921844, + "grad_norm": 2.259274336180325, + "learning_rate": 3.994640738419402e-06, + "loss": 1.155, + "step": 438 + }, + { + "epoch": 0.052786628990560934, + "grad_norm": 2.249835432180165, + "learning_rate": 3.9945835991832075e-06, + "loss": 1.0491, + "step": 439 + }, + { + "epoch": 0.052906871881200024, + "grad_norm": 2.343105787115524, + "learning_rate": 3.994526157370268e-06, + "loss": 1.1714, + "step": 440 + }, + { + "epoch": 0.053027114771839114, + "grad_norm": 0.8871144491543566, + "learning_rate": 3.994468412989296e-06, + "loss": 0.8308, + "step": 441 + }, + { + "epoch": 0.053147357662478203, + "grad_norm": 2.3165529149988138, + "learning_rate": 3.994410366049052e-06, + "loss": 1.1667, + "step": 442 + }, + { + "epoch": 0.0532676005531173, + "grad_norm": 6.985878187391494, + "learning_rate": 3.994352016558341e-06, + "loss": 1.0676, + "step": 443 + }, + { + "epoch": 0.05338784344375639, + "grad_norm": 2.4146138274598896, + "learning_rate": 3.994293364526014e-06, + "loss": 0.9736, + "step": 444 + }, + { + "epoch": 0.05350808633439548, + "grad_norm": 2.7447607242546015, + "learning_rate": 3.99423440996097e-06, + "loss": 1.0894, + "step": 445 + }, + { + "epoch": 0.05362832922503457, + "grad_norm": 14.68489993192811, + "learning_rate": 3.994175152872152e-06, + "loss": 1.0532, + "step": 446 + }, + { + "epoch": 0.05374857211567366, + "grad_norm": 6.77221809507388, + "learning_rate": 3.994115593268548e-06, + "loss": 1.0239, + "step": 447 + }, + { + "epoch": 0.05386881500631275, + "grad_norm": 2.47865596821805, + "learning_rate": 3.994055731159195e-06, + "loss": 1.0572, + "step": 448 + }, + { + "epoch": 0.053989057896951846, + "grad_norm": 2.896137816158462, + "learning_rate": 3.993995566553172e-06, + "loss": 1.1094, + "step": 449 + }, + { + "epoch": 0.054109300787590936, + "grad_norm": 5.428936935252448, + "learning_rate": 3.993935099459607e-06, + "loss": 1.0092, + "step": 450 + }, + { + "epoch": 0.054229543678230026, + "grad_norm": 3.200324346078844, + "learning_rate": 3.993874329887673e-06, + "loss": 0.9805, + "step": 451 + }, + { + "epoch": 0.054349786568869116, + "grad_norm": 4.61652649667813, + "learning_rate": 3.993813257846589e-06, + "loss": 1.1038, + "step": 452 + }, + { + "epoch": 0.054470029459508205, + "grad_norm": 3.079946376036481, + "learning_rate": 3.993751883345619e-06, + "loss": 1.1681, + "step": 453 + }, + { + "epoch": 0.054590272350147295, + "grad_norm": 2.519016704211668, + "learning_rate": 3.993690206394073e-06, + "loss": 1.1133, + "step": 454 + }, + { + "epoch": 0.054710515240786385, + "grad_norm": 2.3733616247348226, + "learning_rate": 3.993628227001307e-06, + "loss": 1.1162, + "step": 455 + }, + { + "epoch": 0.05483075813142548, + "grad_norm": 2.099267096010669, + "learning_rate": 3.993565945176726e-06, + "loss": 0.9513, + "step": 456 + }, + { + "epoch": 0.05495100102206457, + "grad_norm": 2.3892135941582833, + "learning_rate": 3.993503360929776e-06, + "loss": 1.0794, + "step": 457 + }, + { + "epoch": 0.05507124391270366, + "grad_norm": 2.1462234463434884, + "learning_rate": 3.99344047426995e-06, + "loss": 1.0487, + "step": 458 + }, + { + "epoch": 0.05519148680334275, + "grad_norm": 15.717976788108686, + "learning_rate": 3.993377285206789e-06, + "loss": 1.1815, + "step": 459 + }, + { + "epoch": 0.05531172969398184, + "grad_norm": 2.0284477271006, + "learning_rate": 3.99331379374988e-06, + "loss": 1.104, + "step": 460 + }, + { + "epoch": 0.05543197258462093, + "grad_norm": 2.268962007395324, + "learning_rate": 3.993249999908852e-06, + "loss": 1.0381, + "step": 461 + }, + { + "epoch": 0.05555221547526003, + "grad_norm": 2.402809854586114, + "learning_rate": 3.993185903693384e-06, + "loss": 1.1113, + "step": 462 + }, + { + "epoch": 0.05567245836589912, + "grad_norm": 2.5368686059616046, + "learning_rate": 3.9931215051131995e-06, + "loss": 1.0631, + "step": 463 + }, + { + "epoch": 0.05579270125653821, + "grad_norm": 2.3343130734156543, + "learning_rate": 3.993056804178068e-06, + "loss": 1.0399, + "step": 464 + }, + { + "epoch": 0.0559129441471773, + "grad_norm": 2.061859493792813, + "learning_rate": 3.992991800897803e-06, + "loss": 1.0826, + "step": 465 + }, + { + "epoch": 0.05603318703781639, + "grad_norm": 2.580255818510781, + "learning_rate": 3.9929264952822665e-06, + "loss": 1.1396, + "step": 466 + }, + { + "epoch": 0.05615342992845548, + "grad_norm": 13.933300805548217, + "learning_rate": 3.992860887341366e-06, + "loss": 1.1205, + "step": 467 + }, + { + "epoch": 0.056273672819094574, + "grad_norm": 3.0124358870779115, + "learning_rate": 3.992794977085052e-06, + "loss": 1.0503, + "step": 468 + }, + { + "epoch": 0.056393915709733664, + "grad_norm": 4.901678786450323, + "learning_rate": 3.992728764523326e-06, + "loss": 1.087, + "step": 469 + }, + { + "epoch": 0.05651415860037275, + "grad_norm": 4.653259136041985, + "learning_rate": 3.99266224966623e-06, + "loss": 1.0492, + "step": 470 + }, + { + "epoch": 0.05663440149101184, + "grad_norm": 2.3089957621433497, + "learning_rate": 3.992595432523855e-06, + "loss": 1.1194, + "step": 471 + }, + { + "epoch": 0.05675464438165093, + "grad_norm": 2.2722006248291273, + "learning_rate": 3.992528313106338e-06, + "loss": 1.0973, + "step": 472 + }, + { + "epoch": 0.05687488727229002, + "grad_norm": 2.9054011263956703, + "learning_rate": 3.9924608914238595e-06, + "loss": 1.0592, + "step": 473 + }, + { + "epoch": 0.05699513016292912, + "grad_norm": 3.19755315216005, + "learning_rate": 3.992393167486648e-06, + "loss": 1.0796, + "step": 474 + }, + { + "epoch": 0.05711537305356821, + "grad_norm": 2.4718691279634544, + "learning_rate": 3.992325141304977e-06, + "loss": 1.045, + "step": 475 + }, + { + "epoch": 0.0572356159442073, + "grad_norm": 2.320748952796122, + "learning_rate": 3.992256812889166e-06, + "loss": 1.107, + "step": 476 + }, + { + "epoch": 0.05735585883484639, + "grad_norm": 2.6379657424541767, + "learning_rate": 3.992188182249582e-06, + "loss": 1.0087, + "step": 477 + }, + { + "epoch": 0.05747610172548548, + "grad_norm": 2.342722756655337, + "learning_rate": 3.992119249396633e-06, + "loss": 1.1452, + "step": 478 + }, + { + "epoch": 0.05759634461612457, + "grad_norm": 2.6773650222792584, + "learning_rate": 3.992050014340778e-06, + "loss": 1.061, + "step": 479 + }, + { + "epoch": 0.057716587506763666, + "grad_norm": 0.8482324777530952, + "learning_rate": 3.99198047709252e-06, + "loss": 0.7887, + "step": 480 + }, + { + "epoch": 0.057836830397402755, + "grad_norm": 2.276859081379157, + "learning_rate": 3.991910637662408e-06, + "loss": 1.0259, + "step": 481 + }, + { + "epoch": 0.057957073288041845, + "grad_norm": 2.0999494410286905, + "learning_rate": 3.9918404960610355e-06, + "loss": 1.046, + "step": 482 + }, + { + "epoch": 0.058077316178680935, + "grad_norm": 2.5822766716878087, + "learning_rate": 3.991770052299043e-06, + "loss": 1.015, + "step": 483 + }, + { + "epoch": 0.058197559069320025, + "grad_norm": 3.3031871086063065, + "learning_rate": 3.991699306387118e-06, + "loss": 1.1207, + "step": 484 + }, + { + "epoch": 0.058317801959959115, + "grad_norm": 2.1423693601290434, + "learning_rate": 3.991628258335991e-06, + "loss": 1.0193, + "step": 485 + }, + { + "epoch": 0.05843804485059821, + "grad_norm": 3.095033937924477, + "learning_rate": 3.991556908156442e-06, + "loss": 1.1143, + "step": 486 + }, + { + "epoch": 0.0585582877412373, + "grad_norm": 2.6179895794619554, + "learning_rate": 3.9914852558592914e-06, + "loss": 1.1161, + "step": 487 + }, + { + "epoch": 0.05867853063187639, + "grad_norm": 3.867620147239452, + "learning_rate": 3.991413301455413e-06, + "loss": 1.0464, + "step": 488 + }, + { + "epoch": 0.05879877352251548, + "grad_norm": 2.721341268457148, + "learning_rate": 3.991341044955719e-06, + "loss": 1.0148, + "step": 489 + }, + { + "epoch": 0.05891901641315457, + "grad_norm": 2.2316931789305747, + "learning_rate": 3.991268486371172e-06, + "loss": 1.0561, + "step": 490 + }, + { + "epoch": 0.05903925930379366, + "grad_norm": 2.7420596883608663, + "learning_rate": 3.991195625712779e-06, + "loss": 1.1186, + "step": 491 + }, + { + "epoch": 0.05915950219443276, + "grad_norm": 2.801951824822384, + "learning_rate": 3.991122462991592e-06, + "loss": 1.0544, + "step": 492 + }, + { + "epoch": 0.05927974508507185, + "grad_norm": 5.223014870587572, + "learning_rate": 3.991048998218712e-06, + "loss": 1.0496, + "step": 493 + }, + { + "epoch": 0.05939998797571094, + "grad_norm": 14.655733599644543, + "learning_rate": 3.990975231405281e-06, + "loss": 1.0095, + "step": 494 + }, + { + "epoch": 0.05952023086635003, + "grad_norm": 1.9461184750842717, + "learning_rate": 3.990901162562491e-06, + "loss": 1.0281, + "step": 495 + }, + { + "epoch": 0.05964047375698912, + "grad_norm": 2.2044588204247515, + "learning_rate": 3.9908267917015765e-06, + "loss": 1.1461, + "step": 496 + }, + { + "epoch": 0.059760716647628206, + "grad_norm": 2.7585949530716367, + "learning_rate": 3.990752118833821e-06, + "loss": 1.1698, + "step": 497 + }, + { + "epoch": 0.0598809595382673, + "grad_norm": 3.152522388109518, + "learning_rate": 3.990677143970553e-06, + "loss": 1.0192, + "step": 498 + }, + { + "epoch": 0.06000120242890639, + "grad_norm": 2.6642448832708054, + "learning_rate": 3.990601867123144e-06, + "loss": 1.0547, + "step": 499 + }, + { + "epoch": 0.06012144531954548, + "grad_norm": 2.4507364684255424, + "learning_rate": 3.990526288303014e-06, + "loss": 1.0868, + "step": 500 + }, + { + "epoch": 0.06024168821018457, + "grad_norm": 1.961406393337556, + "learning_rate": 3.9904504075216295e-06, + "loss": 1.1475, + "step": 501 + }, + { + "epoch": 0.06036193110082366, + "grad_norm": 4.9653239404389415, + "learning_rate": 3.990374224790501e-06, + "loss": 1.1798, + "step": 502 + }, + { + "epoch": 0.06048217399146275, + "grad_norm": 3.352768616522032, + "learning_rate": 3.990297740121185e-06, + "loss": 0.95, + "step": 503 + }, + { + "epoch": 0.06060241688210185, + "grad_norm": 2.369681879092501, + "learning_rate": 3.990220953525284e-06, + "loss": 1.02, + "step": 504 + }, + { + "epoch": 0.06072265977274094, + "grad_norm": 3.096250198111684, + "learning_rate": 3.9901438650144465e-06, + "loss": 0.9831, + "step": 505 + }, + { + "epoch": 0.06084290266338003, + "grad_norm": 5.8589798043810095, + "learning_rate": 3.990066474600367e-06, + "loss": 1.1547, + "step": 506 + }, + { + "epoch": 0.06096314555401912, + "grad_norm": 2.0631198529604244, + "learning_rate": 3.989988782294786e-06, + "loss": 0.9238, + "step": 507 + }, + { + "epoch": 0.06108338844465821, + "grad_norm": 2.1527198975141473, + "learning_rate": 3.989910788109489e-06, + "loss": 1.1899, + "step": 508 + }, + { + "epoch": 0.0612036313352973, + "grad_norm": 2.422178384790555, + "learning_rate": 3.989832492056307e-06, + "loss": 0.9881, + "step": 509 + }, + { + "epoch": 0.06132387422593639, + "grad_norm": 2.960324347226785, + "learning_rate": 3.989753894147119e-06, + "loss": 1.0442, + "step": 510 + }, + { + "epoch": 0.061444117116575485, + "grad_norm": 1.8474566813592412, + "learning_rate": 3.989674994393846e-06, + "loss": 1.0407, + "step": 511 + }, + { + "epoch": 0.061564360007214575, + "grad_norm": 2.559241945117476, + "learning_rate": 3.98959579280846e-06, + "loss": 1.1784, + "step": 512 + }, + { + "epoch": 0.061684602897853665, + "grad_norm": 2.788703190074254, + "learning_rate": 3.989516289402973e-06, + "loss": 1.0686, + "step": 513 + }, + { + "epoch": 0.061804845788492754, + "grad_norm": 2.940209944618346, + "learning_rate": 3.989436484189447e-06, + "loss": 1.0379, + "step": 514 + }, + { + "epoch": 0.061925088679131844, + "grad_norm": 5.343088566406774, + "learning_rate": 3.9893563771799885e-06, + "loss": 1.0463, + "step": 515 + }, + { + "epoch": 0.062045331569770934, + "grad_norm": 2.9575215869157665, + "learning_rate": 3.989275968386749e-06, + "loss": 1.0991, + "step": 516 + }, + { + "epoch": 0.06216557446041003, + "grad_norm": 2.4074721762375386, + "learning_rate": 3.989195257821926e-06, + "loss": 1.0076, + "step": 517 + }, + { + "epoch": 0.06228581735104912, + "grad_norm": 2.094565161686963, + "learning_rate": 3.989114245497765e-06, + "loss": 1.0809, + "step": 518 + }, + { + "epoch": 0.06240606024168821, + "grad_norm": 2.7529097681857393, + "learning_rate": 3.989032931426554e-06, + "loss": 1.1875, + "step": 519 + }, + { + "epoch": 0.06252630313232731, + "grad_norm": 3.2959620942926757, + "learning_rate": 3.9889513156206295e-06, + "loss": 1.1101, + "step": 520 + }, + { + "epoch": 0.06264654602296639, + "grad_norm": 2.8918380254125866, + "learning_rate": 3.988869398092371e-06, + "loss": 0.9795, + "step": 521 + }, + { + "epoch": 0.06276678891360549, + "grad_norm": 3.9371983246056876, + "learning_rate": 3.988787178854206e-06, + "loss": 1.0217, + "step": 522 + }, + { + "epoch": 0.06288703180424457, + "grad_norm": 2.340713670423608, + "learning_rate": 3.988704657918608e-06, + "loss": 1.1161, + "step": 523 + }, + { + "epoch": 0.06300727469488367, + "grad_norm": 3.548383215653207, + "learning_rate": 3.988621835298094e-06, + "loss": 1.0388, + "step": 524 + }, + { + "epoch": 0.06312751758552275, + "grad_norm": 1.9705302348824016, + "learning_rate": 3.988538711005229e-06, + "loss": 1.1576, + "step": 525 + }, + { + "epoch": 0.06324776047616185, + "grad_norm": 3.1801469550575363, + "learning_rate": 3.988455285052622e-06, + "loss": 1.1193, + "step": 526 + }, + { + "epoch": 0.06336800336680094, + "grad_norm": 2.795015452142629, + "learning_rate": 3.98837155745293e-06, + "loss": 1.0787, + "step": 527 + }, + { + "epoch": 0.06348824625744003, + "grad_norm": 2.741532032761557, + "learning_rate": 3.988287528218854e-06, + "loss": 1.0012, + "step": 528 + }, + { + "epoch": 0.06360848914807912, + "grad_norm": 2.298955895981896, + "learning_rate": 3.98820319736314e-06, + "loss": 1.1443, + "step": 529 + }, + { + "epoch": 0.0637287320387182, + "grad_norm": 1.990218262611883, + "learning_rate": 3.988118564898582e-06, + "loss": 1.091, + "step": 530 + }, + { + "epoch": 0.0638489749293573, + "grad_norm": 2.489398316301224, + "learning_rate": 3.988033630838019e-06, + "loss": 1.1317, + "step": 531 + }, + { + "epoch": 0.0639692178199964, + "grad_norm": 4.514828727236427, + "learning_rate": 3.987948395194334e-06, + "loss": 1.1212, + "step": 532 + }, + { + "epoch": 0.06408946071063548, + "grad_norm": 2.154379434118973, + "learning_rate": 3.987862857980458e-06, + "loss": 1.0057, + "step": 533 + }, + { + "epoch": 0.06420970360127458, + "grad_norm": 2.031694334288738, + "learning_rate": 3.987777019209368e-06, + "loss": 1.007, + "step": 534 + }, + { + "epoch": 0.06432994649191366, + "grad_norm": 2.427219228801112, + "learning_rate": 3.987690878894084e-06, + "loss": 1.0489, + "step": 535 + }, + { + "epoch": 0.06445018938255276, + "grad_norm": 2.4642906958962074, + "learning_rate": 3.987604437047673e-06, + "loss": 1.0949, + "step": 536 + }, + { + "epoch": 0.06457043227319184, + "grad_norm": 2.8488847256949255, + "learning_rate": 3.987517693683251e-06, + "loss": 1.0138, + "step": 537 + }, + { + "epoch": 0.06469067516383094, + "grad_norm": 2.8344189490093763, + "learning_rate": 3.9874306488139745e-06, + "loss": 1.2032, + "step": 538 + }, + { + "epoch": 0.06481091805447003, + "grad_norm": 2.742650528204576, + "learning_rate": 3.987343302453049e-06, + "loss": 1.1198, + "step": 539 + }, + { + "epoch": 0.06493116094510912, + "grad_norm": 1.8562997984708662, + "learning_rate": 3.987255654613724e-06, + "loss": 1.0674, + "step": 540 + }, + { + "epoch": 0.06505140383574821, + "grad_norm": 2.6733304348436024, + "learning_rate": 3.987167705309296e-06, + "loss": 0.9444, + "step": 541 + }, + { + "epoch": 0.0651716467263873, + "grad_norm": 2.3604364091856156, + "learning_rate": 3.987079454553108e-06, + "loss": 1.1882, + "step": 542 + }, + { + "epoch": 0.0652918896170264, + "grad_norm": 1.811498233588987, + "learning_rate": 3.986990902358546e-06, + "loss": 1.15, + "step": 543 + }, + { + "epoch": 0.06541213250766549, + "grad_norm": 2.6903876825781743, + "learning_rate": 3.986902048739045e-06, + "loss": 1.1738, + "step": 544 + }, + { + "epoch": 0.06553237539830457, + "grad_norm": 4.331773232566417, + "learning_rate": 3.986812893708082e-06, + "loss": 1.0452, + "step": 545 + }, + { + "epoch": 0.06565261828894367, + "grad_norm": 3.0504331500049178, + "learning_rate": 3.9867234372791826e-06, + "loss": 1.055, + "step": 546 + }, + { + "epoch": 0.06577286117958275, + "grad_norm": 3.2609319960104934, + "learning_rate": 3.986633679465918e-06, + "loss": 1.1082, + "step": 547 + }, + { + "epoch": 0.06589310407022185, + "grad_norm": 4.048476679151166, + "learning_rate": 3.986543620281904e-06, + "loss": 1.0462, + "step": 548 + }, + { + "epoch": 0.06601334696086093, + "grad_norm": 1.8058513505028355, + "learning_rate": 3.986453259740802e-06, + "loss": 1.1487, + "step": 549 + }, + { + "epoch": 0.06613358985150003, + "grad_norm": 2.7244725128036182, + "learning_rate": 3.986362597856319e-06, + "loss": 1.0307, + "step": 550 + }, + { + "epoch": 0.06625383274213913, + "grad_norm": 3.579765376921872, + "learning_rate": 3.986271634642211e-06, + "loss": 1.054, + "step": 551 + }, + { + "epoch": 0.06637407563277821, + "grad_norm": 2.580638700427151, + "learning_rate": 3.986180370112274e-06, + "loss": 1.0532, + "step": 552 + }, + { + "epoch": 0.0664943185234173, + "grad_norm": 1.9736916421449184, + "learning_rate": 3.986088804280354e-06, + "loss": 0.9837, + "step": 553 + }, + { + "epoch": 0.06661456141405639, + "grad_norm": 3.2776495662405933, + "learning_rate": 3.985996937160342e-06, + "loss": 1.1794, + "step": 554 + }, + { + "epoch": 0.06673480430469549, + "grad_norm": 2.1002391633584123, + "learning_rate": 3.985904768766173e-06, + "loss": 0.9263, + "step": 555 + }, + { + "epoch": 0.06685504719533458, + "grad_norm": 3.460280930905549, + "learning_rate": 3.98581229911183e-06, + "loss": 0.9987, + "step": 556 + }, + { + "epoch": 0.06697529008597367, + "grad_norm": 2.460057952167233, + "learning_rate": 3.985719528211341e-06, + "loss": 1.1585, + "step": 557 + }, + { + "epoch": 0.06709553297661276, + "grad_norm": 0.8592024890105284, + "learning_rate": 3.985626456078777e-06, + "loss": 0.869, + "step": 558 + }, + { + "epoch": 0.06721577586725185, + "grad_norm": 5.323392763389309, + "learning_rate": 3.985533082728259e-06, + "loss": 1.1015, + "step": 559 + }, + { + "epoch": 0.06733601875789094, + "grad_norm": 2.26983735060589, + "learning_rate": 3.985439408173951e-06, + "loss": 0.9855, + "step": 560 + }, + { + "epoch": 0.06745626164853002, + "grad_norm": 2.210378731256162, + "learning_rate": 3.9853454324300634e-06, + "loss": 0.9469, + "step": 561 + }, + { + "epoch": 0.06757650453916912, + "grad_norm": 2.1718479950687346, + "learning_rate": 3.985251155510852e-06, + "loss": 1.0191, + "step": 562 + }, + { + "epoch": 0.06769674742980822, + "grad_norm": 2.606749990242102, + "learning_rate": 3.98515657743062e-06, + "loss": 1.0392, + "step": 563 + }, + { + "epoch": 0.0678169903204473, + "grad_norm": 2.320370143598215, + "learning_rate": 3.985061698203711e-06, + "loss": 1.0166, + "step": 564 + }, + { + "epoch": 0.0679372332110864, + "grad_norm": 0.8959441375220442, + "learning_rate": 3.984966517844523e-06, + "loss": 0.8763, + "step": 565 + }, + { + "epoch": 0.06805747610172548, + "grad_norm": 3.396493219645974, + "learning_rate": 3.984871036367492e-06, + "loss": 1.0467, + "step": 566 + }, + { + "epoch": 0.06817771899236458, + "grad_norm": 2.590454581882473, + "learning_rate": 3.984775253787102e-06, + "loss": 1.0699, + "step": 567 + }, + { + "epoch": 0.06829796188300366, + "grad_norm": 5.753343688024321, + "learning_rate": 3.984679170117885e-06, + "loss": 1.1187, + "step": 568 + }, + { + "epoch": 0.06841820477364276, + "grad_norm": 7.273988172293385, + "learning_rate": 3.984582785374415e-06, + "loss": 1.0241, + "step": 569 + }, + { + "epoch": 0.06853844766428185, + "grad_norm": 2.644004870815667, + "learning_rate": 3.9844860995713155e-06, + "loss": 1.0472, + "step": 570 + }, + { + "epoch": 0.06865869055492094, + "grad_norm": 7.3058822552340486, + "learning_rate": 3.9843891127232524e-06, + "loss": 1.0645, + "step": 571 + }, + { + "epoch": 0.06877893344556003, + "grad_norm": 2.8252741317751333, + "learning_rate": 3.984291824844938e-06, + "loss": 0.9073, + "step": 572 + }, + { + "epoch": 0.06889917633619912, + "grad_norm": 10.065941642605711, + "learning_rate": 3.984194235951132e-06, + "loss": 1.09, + "step": 573 + }, + { + "epoch": 0.06901941922683821, + "grad_norm": 3.334997826479728, + "learning_rate": 3.9840963460566375e-06, + "loss": 1.0899, + "step": 574 + }, + { + "epoch": 0.06913966211747731, + "grad_norm": 1.8569563192962542, + "learning_rate": 3.983998155176305e-06, + "loss": 1.1338, + "step": 575 + }, + { + "epoch": 0.06925990500811639, + "grad_norm": 0.8134334410816387, + "learning_rate": 3.9838996633250305e-06, + "loss": 0.8084, + "step": 576 + }, + { + "epoch": 0.06938014789875549, + "grad_norm": 2.4360295286283713, + "learning_rate": 3.983800870517753e-06, + "loss": 1.124, + "step": 577 + }, + { + "epoch": 0.06950039078939457, + "grad_norm": 5.149400450241585, + "learning_rate": 3.983701776769463e-06, + "loss": 1.0237, + "step": 578 + }, + { + "epoch": 0.06962063368003367, + "grad_norm": 3.9839131405488737, + "learning_rate": 3.9836023820951885e-06, + "loss": 1.1001, + "step": 579 + }, + { + "epoch": 0.06974087657067275, + "grad_norm": 2.367681369944038, + "learning_rate": 3.983502686510011e-06, + "loss": 0.9223, + "step": 580 + }, + { + "epoch": 0.06986111946131185, + "grad_norm": 2.025794699815853, + "learning_rate": 3.9834026900290525e-06, + "loss": 0.9724, + "step": 581 + }, + { + "epoch": 0.06998136235195095, + "grad_norm": 2.9090447882767627, + "learning_rate": 3.983302392667482e-06, + "loss": 1.2436, + "step": 582 + }, + { + "epoch": 0.07010160524259003, + "grad_norm": 1.8820166587659846, + "learning_rate": 3.983201794440517e-06, + "loss": 1.1741, + "step": 583 + }, + { + "epoch": 0.07022184813322913, + "grad_norm": 2.7181515089730413, + "learning_rate": 3.9831008953634165e-06, + "loss": 0.9138, + "step": 584 + }, + { + "epoch": 0.07034209102386821, + "grad_norm": 7.097609490937061, + "learning_rate": 3.9829996954514864e-06, + "loss": 1.0501, + "step": 585 + }, + { + "epoch": 0.0704623339145073, + "grad_norm": 2.189057041245815, + "learning_rate": 3.982898194720079e-06, + "loss": 1.0815, + "step": 586 + }, + { + "epoch": 0.0705825768051464, + "grad_norm": 2.1985213425732306, + "learning_rate": 3.982796393184592e-06, + "loss": 1.0604, + "step": 587 + }, + { + "epoch": 0.07070281969578548, + "grad_norm": 0.7887645196560642, + "learning_rate": 3.98269429086047e-06, + "loss": 0.8638, + "step": 588 + }, + { + "epoch": 0.07082306258642458, + "grad_norm": 2.8278582536712205, + "learning_rate": 3.982591887763199e-06, + "loss": 1.1028, + "step": 589 + }, + { + "epoch": 0.07094330547706366, + "grad_norm": 2.235073487270554, + "learning_rate": 3.982489183908316e-06, + "loss": 1.0522, + "step": 590 + }, + { + "epoch": 0.07106354836770276, + "grad_norm": 1.8394073811155571, + "learning_rate": 3.982386179311399e-06, + "loss": 1.0867, + "step": 591 + }, + { + "epoch": 0.07118379125834184, + "grad_norm": 2.9544564933988653, + "learning_rate": 3.982282873988075e-06, + "loss": 1.1159, + "step": 592 + }, + { + "epoch": 0.07130403414898094, + "grad_norm": 1.8433768331190465, + "learning_rate": 3.982179267954016e-06, + "loss": 1.1078, + "step": 593 + }, + { + "epoch": 0.07142427703962004, + "grad_norm": 2.743066543578555, + "learning_rate": 3.982075361224937e-06, + "loss": 1.2027, + "step": 594 + }, + { + "epoch": 0.07154451993025912, + "grad_norm": 2.1318567391191956, + "learning_rate": 3.981971153816602e-06, + "loss": 1.1194, + "step": 595 + }, + { + "epoch": 0.07166476282089822, + "grad_norm": 1.865086031302456, + "learning_rate": 3.981866645744819e-06, + "loss": 1.2011, + "step": 596 + }, + { + "epoch": 0.0717850057115373, + "grad_norm": 2.5925304809073304, + "learning_rate": 3.9817618370254416e-06, + "loss": 1.0571, + "step": 597 + }, + { + "epoch": 0.0719052486021764, + "grad_norm": 2.6809076584163285, + "learning_rate": 3.9816567276743684e-06, + "loss": 1.1067, + "step": 598 + }, + { + "epoch": 0.0720254914928155, + "grad_norm": 3.041733849885226, + "learning_rate": 3.9815513177075466e-06, + "loss": 1.0137, + "step": 599 + }, + { + "epoch": 0.07214573438345458, + "grad_norm": 1.8480995082032432, + "learning_rate": 3.9814456071409646e-06, + "loss": 0.9412, + "step": 600 + }, + { + "epoch": 0.07226597727409367, + "grad_norm": 2.9532412551521783, + "learning_rate": 3.981339595990659e-06, + "loss": 1.0939, + "step": 601 + }, + { + "epoch": 0.07238622016473276, + "grad_norm": 4.22281524330648, + "learning_rate": 3.981233284272713e-06, + "loss": 1.0515, + "step": 602 + }, + { + "epoch": 0.07250646305537185, + "grad_norm": 1.8228779670603816, + "learning_rate": 3.981126672003253e-06, + "loss": 1.1377, + "step": 603 + }, + { + "epoch": 0.07262670594601094, + "grad_norm": 3.3350434367083506, + "learning_rate": 3.981019759198451e-06, + "loss": 1.0234, + "step": 604 + }, + { + "epoch": 0.07274694883665003, + "grad_norm": 2.6043910324082904, + "learning_rate": 3.980912545874528e-06, + "loss": 1.0828, + "step": 605 + }, + { + "epoch": 0.07286719172728913, + "grad_norm": 3.387207277607837, + "learning_rate": 3.980805032047746e-06, + "loss": 1.0945, + "step": 606 + }, + { + "epoch": 0.07298743461792821, + "grad_norm": 2.183326445512851, + "learning_rate": 3.980697217734415e-06, + "loss": 1.0507, + "step": 607 + }, + { + "epoch": 0.07310767750856731, + "grad_norm": 3.4318798939245005, + "learning_rate": 3.980589102950891e-06, + "loss": 1.157, + "step": 608 + }, + { + "epoch": 0.07322792039920639, + "grad_norm": 4.585262840333301, + "learning_rate": 3.9804806877135755e-06, + "loss": 0.996, + "step": 609 + }, + { + "epoch": 0.07334816328984549, + "grad_norm": 2.5236961775430515, + "learning_rate": 3.980371972038915e-06, + "loss": 1.1005, + "step": 610 + }, + { + "epoch": 0.07346840618048459, + "grad_norm": 1.9957453055935228, + "learning_rate": 3.980262955943399e-06, + "loss": 1.083, + "step": 611 + }, + { + "epoch": 0.07358864907112367, + "grad_norm": 2.6649476312068865, + "learning_rate": 3.980153639443569e-06, + "loss": 1.1075, + "step": 612 + }, + { + "epoch": 0.07370889196176277, + "grad_norm": 2.265404841808642, + "learning_rate": 3.980044022556005e-06, + "loss": 1.0386, + "step": 613 + }, + { + "epoch": 0.07382913485240185, + "grad_norm": 3.2461689597183727, + "learning_rate": 3.9799341052973375e-06, + "loss": 0.9699, + "step": 614 + }, + { + "epoch": 0.07394937774304094, + "grad_norm": 10.019782526088937, + "learning_rate": 3.979823887684241e-06, + "loss": 0.9931, + "step": 615 + }, + { + "epoch": 0.07406962063368003, + "grad_norm": 3.2723867692001702, + "learning_rate": 3.979713369733434e-06, + "loss": 1.0839, + "step": 616 + }, + { + "epoch": 0.07418986352431912, + "grad_norm": 2.3919345036728825, + "learning_rate": 3.979602551461683e-06, + "loss": 1.0911, + "step": 617 + }, + { + "epoch": 0.07431010641495822, + "grad_norm": 4.808709571614886, + "learning_rate": 3.979491432885799e-06, + "loss": 1.153, + "step": 618 + }, + { + "epoch": 0.0744303493055973, + "grad_norm": 2.651815055549061, + "learning_rate": 3.97938001402264e-06, + "loss": 1.0701, + "step": 619 + }, + { + "epoch": 0.0745505921962364, + "grad_norm": 4.503950504241063, + "learning_rate": 3.979268294889105e-06, + "loss": 1.0429, + "step": 620 + }, + { + "epoch": 0.07467083508687548, + "grad_norm": 1.9559886233695165, + "learning_rate": 3.979156275502143e-06, + "loss": 0.9786, + "step": 621 + }, + { + "epoch": 0.07479107797751458, + "grad_norm": 2.74437050721946, + "learning_rate": 3.979043955878749e-06, + "loss": 1.158, + "step": 622 + }, + { + "epoch": 0.07491132086815366, + "grad_norm": 4.571137495651426, + "learning_rate": 3.978931336035959e-06, + "loss": 1.0709, + "step": 623 + }, + { + "epoch": 0.07503156375879276, + "grad_norm": 2.570694080540331, + "learning_rate": 3.9788184159908595e-06, + "loss": 1.0637, + "step": 624 + }, + { + "epoch": 0.07515180664943186, + "grad_norm": 2.943268165111744, + "learning_rate": 3.97870519576058e-06, + "loss": 1.0612, + "step": 625 + }, + { + "epoch": 0.07527204954007094, + "grad_norm": 2.558193029808108, + "learning_rate": 3.978591675362295e-06, + "loss": 1.0511, + "step": 626 + }, + { + "epoch": 0.07539229243071004, + "grad_norm": 1.9040214143718863, + "learning_rate": 3.978477854813226e-06, + "loss": 1.1127, + "step": 627 + }, + { + "epoch": 0.07551253532134912, + "grad_norm": 2.5282628836358674, + "learning_rate": 3.97836373413064e-06, + "loss": 1.0637, + "step": 628 + }, + { + "epoch": 0.07563277821198822, + "grad_norm": 7.326810999597855, + "learning_rate": 3.978249313331848e-06, + "loss": 0.9821, + "step": 629 + }, + { + "epoch": 0.07575302110262731, + "grad_norm": 4.22008126058578, + "learning_rate": 3.978134592434208e-06, + "loss": 0.8608, + "step": 630 + }, + { + "epoch": 0.0758732639932664, + "grad_norm": 1.0294455144061985, + "learning_rate": 3.978019571455123e-06, + "loss": 0.8319, + "step": 631 + }, + { + "epoch": 0.07599350688390549, + "grad_norm": 3.7330674951001934, + "learning_rate": 3.977904250412042e-06, + "loss": 1.0812, + "step": 632 + }, + { + "epoch": 0.07611374977454458, + "grad_norm": 2.638897934836783, + "learning_rate": 3.97778862932246e-06, + "loss": 1.0939, + "step": 633 + }, + { + "epoch": 0.07623399266518367, + "grad_norm": 2.5741200174896504, + "learning_rate": 3.9776727082039144e-06, + "loss": 1.1784, + "step": 634 + }, + { + "epoch": 0.07635423555582276, + "grad_norm": 0.8028099599926519, + "learning_rate": 3.977556487073991e-06, + "loss": 0.7926, + "step": 635 + }, + { + "epoch": 0.07647447844646185, + "grad_norm": 2.3120890071723887, + "learning_rate": 3.97743996595032e-06, + "loss": 1.0558, + "step": 636 + }, + { + "epoch": 0.07659472133710095, + "grad_norm": 1.7268571559054489, + "learning_rate": 3.9773231448505804e-06, + "loss": 1.055, + "step": 637 + }, + { + "epoch": 0.07671496422774003, + "grad_norm": 2.977658207787587, + "learning_rate": 3.977206023792491e-06, + "loss": 1.0183, + "step": 638 + }, + { + "epoch": 0.07683520711837913, + "grad_norm": 2.8666476123383555, + "learning_rate": 3.97708860279382e-06, + "loss": 1.0489, + "step": 639 + }, + { + "epoch": 0.07695545000901821, + "grad_norm": 1.8596960479791156, + "learning_rate": 3.97697088187238e-06, + "loss": 1.022, + "step": 640 + }, + { + "epoch": 0.07707569289965731, + "grad_norm": 2.670694279931556, + "learning_rate": 3.976852861046029e-06, + "loss": 1.1571, + "step": 641 + }, + { + "epoch": 0.0771959357902964, + "grad_norm": 1.7690329281603345, + "learning_rate": 3.97673454033267e-06, + "loss": 1.0399, + "step": 642 + }, + { + "epoch": 0.07731617868093549, + "grad_norm": 3.271388174891557, + "learning_rate": 3.976615919750254e-06, + "loss": 1.0676, + "step": 643 + }, + { + "epoch": 0.07743642157157458, + "grad_norm": 2.610894406121119, + "learning_rate": 3.976496999316775e-06, + "loss": 1.1042, + "step": 644 + }, + { + "epoch": 0.07755666446221367, + "grad_norm": 2.3761199519065506, + "learning_rate": 3.976377779050271e-06, + "loss": 1.0783, + "step": 645 + }, + { + "epoch": 0.07767690735285276, + "grad_norm": 3.0867674665147655, + "learning_rate": 3.976258258968831e-06, + "loss": 1.085, + "step": 646 + }, + { + "epoch": 0.07779715024349185, + "grad_norm": 2.6678545412162897, + "learning_rate": 3.976138439090583e-06, + "loss": 0.9811, + "step": 647 + }, + { + "epoch": 0.07791739313413094, + "grad_norm": 2.413185231826237, + "learning_rate": 3.976018319433706e-06, + "loss": 1.0866, + "step": 648 + }, + { + "epoch": 0.07803763602477004, + "grad_norm": 2.597391077718951, + "learning_rate": 3.9758979000164205e-06, + "loss": 1.1531, + "step": 649 + }, + { + "epoch": 0.07815787891540912, + "grad_norm": 1.9191061403859393, + "learning_rate": 3.975777180856995e-06, + "loss": 0.9576, + "step": 650 + }, + { + "epoch": 0.07827812180604822, + "grad_norm": 3.075249582722143, + "learning_rate": 3.975656161973742e-06, + "loss": 1.1004, + "step": 651 + }, + { + "epoch": 0.0783983646966873, + "grad_norm": 2.714813627446422, + "learning_rate": 3.9755348433850194e-06, + "loss": 1.1268, + "step": 652 + }, + { + "epoch": 0.0785186075873264, + "grad_norm": 0.9595773887256897, + "learning_rate": 3.975413225109232e-06, + "loss": 0.8735, + "step": 653 + }, + { + "epoch": 0.0786388504779655, + "grad_norm": 4.529999238070999, + "learning_rate": 3.975291307164829e-06, + "loss": 1.1798, + "step": 654 + }, + { + "epoch": 0.07875909336860458, + "grad_norm": 3.3191547843790015, + "learning_rate": 3.975169089570306e-06, + "loss": 1.0855, + "step": 655 + }, + { + "epoch": 0.07887933625924368, + "grad_norm": 2.2897112971730422, + "learning_rate": 3.975046572344202e-06, + "loss": 1.1534, + "step": 656 + }, + { + "epoch": 0.07899957914988276, + "grad_norm": 2.126784865324371, + "learning_rate": 3.974923755505103e-06, + "loss": 0.9466, + "step": 657 + }, + { + "epoch": 0.07911982204052186, + "grad_norm": 1.8873970913149027, + "learning_rate": 3.974800639071641e-06, + "loss": 1.1499, + "step": 658 + }, + { + "epoch": 0.07924006493116094, + "grad_norm": 2.3988627202841437, + "learning_rate": 3.974677223062492e-06, + "loss": 1.2442, + "step": 659 + }, + { + "epoch": 0.07936030782180004, + "grad_norm": 2.398856061267421, + "learning_rate": 3.974553507496378e-06, + "loss": 0.983, + "step": 660 + }, + { + "epoch": 0.07948055071243913, + "grad_norm": 2.235247248344234, + "learning_rate": 3.974429492392068e-06, + "loss": 1.134, + "step": 661 + }, + { + "epoch": 0.07960079360307822, + "grad_norm": 2.573214206145304, + "learning_rate": 3.974305177768373e-06, + "loss": 1.1478, + "step": 662 + }, + { + "epoch": 0.07972103649371731, + "grad_norm": 4.005686770644529, + "learning_rate": 3.974180563644152e-06, + "loss": 1.1061, + "step": 663 + }, + { + "epoch": 0.0798412793843564, + "grad_norm": 2.3316628630004717, + "learning_rate": 3.97405565003831e-06, + "loss": 1.1335, + "step": 664 + }, + { + "epoch": 0.07996152227499549, + "grad_norm": 2.3713190766661065, + "learning_rate": 3.973930436969794e-06, + "loss": 1.0203, + "step": 665 + }, + { + "epoch": 0.08008176516563459, + "grad_norm": 1.9221962761031315, + "learning_rate": 3.973804924457602e-06, + "loss": 1.0965, + "step": 666 + }, + { + "epoch": 0.08020200805627367, + "grad_norm": 2.0573795219365665, + "learning_rate": 3.973679112520771e-06, + "loss": 1.095, + "step": 667 + }, + { + "epoch": 0.08032225094691277, + "grad_norm": 2.5514292806594785, + "learning_rate": 3.973553001178389e-06, + "loss": 1.2276, + "step": 668 + }, + { + "epoch": 0.08044249383755185, + "grad_norm": 2.302508238737296, + "learning_rate": 3.973426590449585e-06, + "loss": 0.9921, + "step": 669 + }, + { + "epoch": 0.08056273672819095, + "grad_norm": 2.8134881720788325, + "learning_rate": 3.9732998803535364e-06, + "loss": 0.9903, + "step": 670 + }, + { + "epoch": 0.08068297961883003, + "grad_norm": 2.4272434907606493, + "learning_rate": 3.973172870909465e-06, + "loss": 1.0916, + "step": 671 + }, + { + "epoch": 0.08080322250946913, + "grad_norm": 2.8590601346448845, + "learning_rate": 3.973045562136638e-06, + "loss": 1.0473, + "step": 672 + }, + { + "epoch": 0.08092346540010822, + "grad_norm": 2.946674224389282, + "learning_rate": 3.972917954054368e-06, + "loss": 1.1533, + "step": 673 + }, + { + "epoch": 0.08104370829074731, + "grad_norm": 2.766886191870021, + "learning_rate": 3.972790046682013e-06, + "loss": 1.0617, + "step": 674 + }, + { + "epoch": 0.0811639511813864, + "grad_norm": 1.8830953678366043, + "learning_rate": 3.972661840038977e-06, + "loss": 1.029, + "step": 675 + }, + { + "epoch": 0.08128419407202549, + "grad_norm": 3.0450158340217803, + "learning_rate": 3.972533334144707e-06, + "loss": 1.0777, + "step": 676 + }, + { + "epoch": 0.08140443696266458, + "grad_norm": 4.587505222078517, + "learning_rate": 3.972404529018699e-06, + "loss": 1.0282, + "step": 677 + }, + { + "epoch": 0.08152467985330367, + "grad_norm": 2.6131349075009127, + "learning_rate": 3.972275424680493e-06, + "loss": 1.0927, + "step": 678 + }, + { + "epoch": 0.08164492274394276, + "grad_norm": 2.209102016996178, + "learning_rate": 3.972146021149673e-06, + "loss": 1.1569, + "step": 679 + }, + { + "epoch": 0.08176516563458186, + "grad_norm": 3.1226662122898223, + "learning_rate": 3.972016318445868e-06, + "loss": 1.0257, + "step": 680 + }, + { + "epoch": 0.08188540852522094, + "grad_norm": 2.3776731004084657, + "learning_rate": 3.971886316588757e-06, + "loss": 1.1624, + "step": 681 + }, + { + "epoch": 0.08200565141586004, + "grad_norm": 6.146057676541842, + "learning_rate": 3.9717560155980595e-06, + "loss": 0.9702, + "step": 682 + }, + { + "epoch": 0.08212589430649912, + "grad_norm": 2.594009386444168, + "learning_rate": 3.971625415493542e-06, + "loss": 1.1592, + "step": 683 + }, + { + "epoch": 0.08224613719713822, + "grad_norm": 2.1559542024592746, + "learning_rate": 3.971494516295017e-06, + "loss": 1.1187, + "step": 684 + }, + { + "epoch": 0.08236638008777732, + "grad_norm": 1.9386558776478295, + "learning_rate": 3.971363318022341e-06, + "loss": 1.0944, + "step": 685 + }, + { + "epoch": 0.0824866229784164, + "grad_norm": 2.295897606736345, + "learning_rate": 3.971231820695417e-06, + "loss": 0.9247, + "step": 686 + }, + { + "epoch": 0.0826068658690555, + "grad_norm": 1.9933585521076589, + "learning_rate": 3.971100024334193e-06, + "loss": 1.0542, + "step": 687 + }, + { + "epoch": 0.08272710875969458, + "grad_norm": 2.652742830571121, + "learning_rate": 3.970967928958663e-06, + "loss": 1.1049, + "step": 688 + }, + { + "epoch": 0.08284735165033368, + "grad_norm": 1.78200896306465, + "learning_rate": 3.970835534588865e-06, + "loss": 1.0732, + "step": 689 + }, + { + "epoch": 0.08296759454097276, + "grad_norm": 2.2875006753889244, + "learning_rate": 3.970702841244883e-06, + "loss": 1.0975, + "step": 690 + }, + { + "epoch": 0.08308783743161186, + "grad_norm": 1.774289838004207, + "learning_rate": 3.970569848946847e-06, + "loss": 1.0641, + "step": 691 + }, + { + "epoch": 0.08320808032225095, + "grad_norm": 4.250625339951717, + "learning_rate": 3.970436557714932e-06, + "loss": 1.0728, + "step": 692 + }, + { + "epoch": 0.08332832321289003, + "grad_norm": 3.0016162015607804, + "learning_rate": 3.970302967569358e-06, + "loss": 1.1052, + "step": 693 + }, + { + "epoch": 0.08344856610352913, + "grad_norm": 2.1789993785981565, + "learning_rate": 3.9701690785303896e-06, + "loss": 0.9205, + "step": 694 + }, + { + "epoch": 0.08356880899416821, + "grad_norm": 2.4705566362823204, + "learning_rate": 3.970034890618339e-06, + "loss": 1.1221, + "step": 695 + }, + { + "epoch": 0.08368905188480731, + "grad_norm": 2.161079968040373, + "learning_rate": 3.969900403853562e-06, + "loss": 1.1197, + "step": 696 + }, + { + "epoch": 0.08380929477544641, + "grad_norm": 2.7811855351552928, + "learning_rate": 3.96976561825646e-06, + "loss": 1.0179, + "step": 697 + }, + { + "epoch": 0.08392953766608549, + "grad_norm": 2.5719301646586565, + "learning_rate": 3.969630533847479e-06, + "loss": 1.1104, + "step": 698 + }, + { + "epoch": 0.08404978055672459, + "grad_norm": 2.1905019597695854, + "learning_rate": 3.969495150647113e-06, + "loss": 1.0811, + "step": 699 + }, + { + "epoch": 0.08417002344736367, + "grad_norm": 2.292615509118384, + "learning_rate": 3.969359468675899e-06, + "loss": 1.0051, + "step": 700 + }, + { + "epoch": 0.08429026633800277, + "grad_norm": 2.8645359844571687, + "learning_rate": 3.969223487954418e-06, + "loss": 1.1352, + "step": 701 + }, + { + "epoch": 0.08441050922864185, + "grad_norm": 2.0936822453254145, + "learning_rate": 3.969087208503301e-06, + "loss": 1.0661, + "step": 702 + }, + { + "epoch": 0.08453075211928095, + "grad_norm": 3.347830733355804, + "learning_rate": 3.968950630343219e-06, + "loss": 1.0845, + "step": 703 + }, + { + "epoch": 0.08465099500992004, + "grad_norm": 2.283902859721499, + "learning_rate": 3.968813753494892e-06, + "loss": 1.1734, + "step": 704 + }, + { + "epoch": 0.08477123790055913, + "grad_norm": 3.0839774172162646, + "learning_rate": 3.968676577979084e-06, + "loss": 0.9965, + "step": 705 + }, + { + "epoch": 0.08489148079119822, + "grad_norm": 2.4979057122998625, + "learning_rate": 3.968539103816605e-06, + "loss": 1.022, + "step": 706 + }, + { + "epoch": 0.0850117236818373, + "grad_norm": 2.039058407600359, + "learning_rate": 3.9684013310283085e-06, + "loss": 1.1315, + "step": 707 + }, + { + "epoch": 0.0851319665724764, + "grad_norm": 3.772549416109067, + "learning_rate": 3.9682632596350956e-06, + "loss": 0.8823, + "step": 708 + }, + { + "epoch": 0.0852522094631155, + "grad_norm": 2.4409397015154486, + "learning_rate": 3.968124889657911e-06, + "loss": 1.0267, + "step": 709 + }, + { + "epoch": 0.08537245235375458, + "grad_norm": 2.587799047613171, + "learning_rate": 3.967986221117746e-06, + "loss": 1.1445, + "step": 710 + }, + { + "epoch": 0.08549269524439368, + "grad_norm": 2.213827663720033, + "learning_rate": 3.967847254035635e-06, + "loss": 1.105, + "step": 711 + }, + { + "epoch": 0.08561293813503276, + "grad_norm": 2.1263443677302654, + "learning_rate": 3.967707988432661e-06, + "loss": 1.1036, + "step": 712 + }, + { + "epoch": 0.08573318102567186, + "grad_norm": 3.490618803587411, + "learning_rate": 3.967568424329949e-06, + "loss": 1.1168, + "step": 713 + }, + { + "epoch": 0.08585342391631094, + "grad_norm": 0.8255543666892079, + "learning_rate": 3.967428561748671e-06, + "loss": 0.7936, + "step": 714 + }, + { + "epoch": 0.08597366680695004, + "grad_norm": 2.1720122149916086, + "learning_rate": 3.967288400710045e-06, + "loss": 1.1179, + "step": 715 + }, + { + "epoch": 0.08609390969758914, + "grad_norm": 2.021362772799339, + "learning_rate": 3.9671479412353335e-06, + "loss": 1.124, + "step": 716 + }, + { + "epoch": 0.08621415258822822, + "grad_norm": 6.676778880620739, + "learning_rate": 3.967007183345843e-06, + "loss": 0.9789, + "step": 717 + }, + { + "epoch": 0.08633439547886732, + "grad_norm": 4.172556617860091, + "learning_rate": 3.966866127062927e-06, + "loss": 1.1312, + "step": 718 + }, + { + "epoch": 0.0864546383695064, + "grad_norm": 0.8740996087836583, + "learning_rate": 3.966724772407982e-06, + "loss": 0.8651, + "step": 719 + }, + { + "epoch": 0.0865748812601455, + "grad_norm": 2.427142021113622, + "learning_rate": 3.966583119402454e-06, + "loss": 1.1268, + "step": 720 + }, + { + "epoch": 0.08669512415078459, + "grad_norm": 1.8199664920824383, + "learning_rate": 3.9664411680678305e-06, + "loss": 1.0585, + "step": 721 + }, + { + "epoch": 0.08681536704142367, + "grad_norm": 0.843521581743397, + "learning_rate": 3.966298918425644e-06, + "loss": 0.8521, + "step": 722 + }, + { + "epoch": 0.08693560993206277, + "grad_norm": 1.9849829155597503, + "learning_rate": 3.966156370497476e-06, + "loss": 1.0636, + "step": 723 + }, + { + "epoch": 0.08705585282270185, + "grad_norm": 2.1436463355977313, + "learning_rate": 3.96601352430495e-06, + "loss": 1.1271, + "step": 724 + }, + { + "epoch": 0.08717609571334095, + "grad_norm": 1.6739105952097966, + "learning_rate": 3.965870379869735e-06, + "loss": 1.0683, + "step": 725 + }, + { + "epoch": 0.08729633860398003, + "grad_norm": 2.39909925247072, + "learning_rate": 3.965726937213547e-06, + "loss": 1.1061, + "step": 726 + }, + { + "epoch": 0.08741658149461913, + "grad_norm": 2.438065389937598, + "learning_rate": 3.965583196358144e-06, + "loss": 1.0532, + "step": 727 + }, + { + "epoch": 0.08753682438525823, + "grad_norm": 2.5877046107970822, + "learning_rate": 3.965439157325335e-06, + "loss": 0.9823, + "step": 728 + }, + { + "epoch": 0.08765706727589731, + "grad_norm": 2.7494104793866954, + "learning_rate": 3.965294820136968e-06, + "loss": 1.0001, + "step": 729 + }, + { + "epoch": 0.08777731016653641, + "grad_norm": 2.259467445687713, + "learning_rate": 3.965150184814938e-06, + "loss": 1.1114, + "step": 730 + }, + { + "epoch": 0.08789755305717549, + "grad_norm": 4.103218136824941, + "learning_rate": 3.965005251381189e-06, + "loss": 1.0049, + "step": 731 + }, + { + "epoch": 0.08801779594781459, + "grad_norm": 11.769235652195688, + "learning_rate": 3.964860019857705e-06, + "loss": 0.8844, + "step": 732 + }, + { + "epoch": 0.08813803883845367, + "grad_norm": 2.4933152472338156, + "learning_rate": 3.964714490266518e-06, + "loss": 1.0804, + "step": 733 + }, + { + "epoch": 0.08825828172909277, + "grad_norm": 0.9214756310708272, + "learning_rate": 3.964568662629706e-06, + "loss": 0.8839, + "step": 734 + }, + { + "epoch": 0.08837852461973186, + "grad_norm": 2.2013199357854094, + "learning_rate": 3.9644225369693895e-06, + "loss": 1.0823, + "step": 735 + }, + { + "epoch": 0.08849876751037095, + "grad_norm": 3.257143825636029, + "learning_rate": 3.964276113307735e-06, + "loss": 1.1068, + "step": 736 + }, + { + "epoch": 0.08861901040101004, + "grad_norm": 2.3469279805183088, + "learning_rate": 3.9641293916669574e-06, + "loss": 1.0491, + "step": 737 + }, + { + "epoch": 0.08873925329164913, + "grad_norm": 3.2905073095940978, + "learning_rate": 3.9639823720693115e-06, + "loss": 1.0681, + "step": 738 + }, + { + "epoch": 0.08885949618228822, + "grad_norm": 0.8334313732332527, + "learning_rate": 3.963835054537102e-06, + "loss": 0.8382, + "step": 739 + }, + { + "epoch": 0.08897973907292732, + "grad_norm": 3.7638363210863344, + "learning_rate": 3.963687439092676e-06, + "loss": 0.8475, + "step": 740 + }, + { + "epoch": 0.0890999819635664, + "grad_norm": 3.254091328535615, + "learning_rate": 3.963539525758427e-06, + "loss": 1.0442, + "step": 741 + }, + { + "epoch": 0.0892202248542055, + "grad_norm": 2.7579024951288504, + "learning_rate": 3.9633913145567925e-06, + "loss": 0.9147, + "step": 742 + }, + { + "epoch": 0.08934046774484458, + "grad_norm": 2.372944695427471, + "learning_rate": 3.9632428055102575e-06, + "loss": 1.0542, + "step": 743 + }, + { + "epoch": 0.08946071063548368, + "grad_norm": 3.2960234351297197, + "learning_rate": 3.9630939986413495e-06, + "loss": 0.9105, + "step": 744 + }, + { + "epoch": 0.08958095352612276, + "grad_norm": 2.0794462435195338, + "learning_rate": 3.962944893972643e-06, + "loss": 1.021, + "step": 745 + }, + { + "epoch": 0.08970119641676186, + "grad_norm": 2.6613625187557206, + "learning_rate": 3.962795491526756e-06, + "loss": 1.1524, + "step": 746 + }, + { + "epoch": 0.08982143930740095, + "grad_norm": 2.906565987421135, + "learning_rate": 3.962645791326354e-06, + "loss": 1.1349, + "step": 747 + }, + { + "epoch": 0.08994168219804004, + "grad_norm": 2.03229343317119, + "learning_rate": 3.962495793394146e-06, + "loss": 1.0686, + "step": 748 + }, + { + "epoch": 0.09006192508867913, + "grad_norm": 0.7405729299662839, + "learning_rate": 3.9623454977528864e-06, + "loss": 0.8503, + "step": 749 + }, + { + "epoch": 0.09018216797931822, + "grad_norm": 2.124338159154733, + "learning_rate": 3.962194904425375e-06, + "loss": 1.088, + "step": 750 + }, + { + "epoch": 0.09030241086995731, + "grad_norm": 4.127625987800496, + "learning_rate": 3.9620440134344566e-06, + "loss": 0.9211, + "step": 751 + }, + { + "epoch": 0.09042265376059641, + "grad_norm": 2.475654122581699, + "learning_rate": 3.9618928248030215e-06, + "loss": 1.0609, + "step": 752 + }, + { + "epoch": 0.0905428966512355, + "grad_norm": 2.4685185963789995, + "learning_rate": 3.961741338554005e-06, + "loss": 1.0681, + "step": 753 + }, + { + "epoch": 0.09066313954187459, + "grad_norm": 2.2300306648232993, + "learning_rate": 3.9615895547103865e-06, + "loss": 0.9989, + "step": 754 + }, + { + "epoch": 0.09078338243251367, + "grad_norm": 2.242431095727513, + "learning_rate": 3.961437473295193e-06, + "loss": 1.0171, + "step": 755 + }, + { + "epoch": 0.09090362532315277, + "grad_norm": 2.403557497152993, + "learning_rate": 3.961285094331495e-06, + "loss": 0.9651, + "step": 756 + }, + { + "epoch": 0.09102386821379185, + "grad_norm": 1.9309385622623236, + "learning_rate": 3.961132417842406e-06, + "loss": 1.0994, + "step": 757 + }, + { + "epoch": 0.09114411110443095, + "grad_norm": 3.0820238236602546, + "learning_rate": 3.960979443851089e-06, + "loss": 0.9912, + "step": 758 + }, + { + "epoch": 0.09126435399507005, + "grad_norm": 2.2609540473422087, + "learning_rate": 3.96082617238075e-06, + "loss": 1.0277, + "step": 759 + }, + { + "epoch": 0.09138459688570913, + "grad_norm": 10.300776435697585, + "learning_rate": 3.960672603454639e-06, + "loss": 1.0378, + "step": 760 + }, + { + "epoch": 0.09150483977634823, + "grad_norm": 3.1545898301459037, + "learning_rate": 3.960518737096054e-06, + "loss": 1.0107, + "step": 761 + }, + { + "epoch": 0.09162508266698731, + "grad_norm": 2.795142633555387, + "learning_rate": 3.960364573328334e-06, + "loss": 0.9681, + "step": 762 + }, + { + "epoch": 0.0917453255576264, + "grad_norm": 2.006434817729051, + "learning_rate": 3.9602101121748675e-06, + "loss": 1.1269, + "step": 763 + }, + { + "epoch": 0.0918655684482655, + "grad_norm": 2.271675711893875, + "learning_rate": 3.960055353659085e-06, + "loss": 0.9667, + "step": 764 + }, + { + "epoch": 0.09198581133890459, + "grad_norm": 2.217898030646793, + "learning_rate": 3.959900297804465e-06, + "loss": 1.0796, + "step": 765 + }, + { + "epoch": 0.09210605422954368, + "grad_norm": 2.2076328774159335, + "learning_rate": 3.9597449446345276e-06, + "loss": 1.0116, + "step": 766 + }, + { + "epoch": 0.09222629712018277, + "grad_norm": 2.7348417919086323, + "learning_rate": 3.95958929417284e-06, + "loss": 1.0677, + "step": 767 + }, + { + "epoch": 0.09234654001082186, + "grad_norm": 0.7407774128600271, + "learning_rate": 3.9594333464430145e-06, + "loss": 0.8262, + "step": 768 + }, + { + "epoch": 0.09246678290146094, + "grad_norm": 2.965646346149056, + "learning_rate": 3.959277101468709e-06, + "loss": 1.123, + "step": 769 + }, + { + "epoch": 0.09258702579210004, + "grad_norm": 2.9257409942261483, + "learning_rate": 3.959120559273624e-06, + "loss": 1.0241, + "step": 770 + }, + { + "epoch": 0.09270726868273914, + "grad_norm": 1.9075211211068863, + "learning_rate": 3.958963719881509e-06, + "loss": 1.0739, + "step": 771 + }, + { + "epoch": 0.09282751157337822, + "grad_norm": 2.0506139487189947, + "learning_rate": 3.958806583316154e-06, + "loss": 1.1763, + "step": 772 + }, + { + "epoch": 0.09294775446401732, + "grad_norm": 2.1407992995088514, + "learning_rate": 3.9586491496013985e-06, + "loss": 1.0261, + "step": 773 + }, + { + "epoch": 0.0930679973546564, + "grad_norm": 3.021594528232071, + "learning_rate": 3.958491418761124e-06, + "loss": 1.0714, + "step": 774 + }, + { + "epoch": 0.0931882402452955, + "grad_norm": 2.970042057959753, + "learning_rate": 3.958333390819258e-06, + "loss": 0.9692, + "step": 775 + }, + { + "epoch": 0.0933084831359346, + "grad_norm": 2.2056837441287382, + "learning_rate": 3.9581750657997754e-06, + "loss": 1.0391, + "step": 776 + }, + { + "epoch": 0.09342872602657368, + "grad_norm": 2.8223354872138207, + "learning_rate": 3.95801644372669e-06, + "loss": 1.1332, + "step": 777 + }, + { + "epoch": 0.09354896891721277, + "grad_norm": 2.1699472558076294, + "learning_rate": 3.957857524624068e-06, + "loss": 1.0838, + "step": 778 + }, + { + "epoch": 0.09366921180785186, + "grad_norm": 1.7633791793651488, + "learning_rate": 3.957698308516016e-06, + "loss": 1.1348, + "step": 779 + }, + { + "epoch": 0.09378945469849095, + "grad_norm": 2.161868198248427, + "learning_rate": 3.957538795426688e-06, + "loss": 1.0628, + "step": 780 + }, + { + "epoch": 0.09390969758913004, + "grad_norm": 2.4193946196249114, + "learning_rate": 3.9573789853802804e-06, + "loss": 1.0126, + "step": 781 + }, + { + "epoch": 0.09402994047976913, + "grad_norm": 2.3903290808540594, + "learning_rate": 3.957218878401037e-06, + "loss": 0.9882, + "step": 782 + }, + { + "epoch": 0.09415018337040823, + "grad_norm": 2.470885885965275, + "learning_rate": 3.957058474513246e-06, + "loss": 1.1321, + "step": 783 + }, + { + "epoch": 0.09427042626104731, + "grad_norm": 1.8363421704327845, + "learning_rate": 3.956897773741241e-06, + "loss": 1.0209, + "step": 784 + }, + { + "epoch": 0.09439066915168641, + "grad_norm": 2.343119422008502, + "learning_rate": 3.956736776109398e-06, + "loss": 0.957, + "step": 785 + }, + { + "epoch": 0.09451091204232549, + "grad_norm": 1.9298783180934218, + "learning_rate": 3.956575481642143e-06, + "loss": 1.0758, + "step": 786 + }, + { + "epoch": 0.09463115493296459, + "grad_norm": 3.9213277123350525, + "learning_rate": 3.956413890363943e-06, + "loss": 0.9918, + "step": 787 + }, + { + "epoch": 0.09475139782360369, + "grad_norm": 2.3844404819041225, + "learning_rate": 3.956252002299312e-06, + "loss": 1.0615, + "step": 788 + }, + { + "epoch": 0.09487164071424277, + "grad_norm": 2.070250772959511, + "learning_rate": 3.956089817472807e-06, + "loss": 1.1444, + "step": 789 + }, + { + "epoch": 0.09499188360488187, + "grad_norm": 2.949138230999353, + "learning_rate": 3.955927335909032e-06, + "loss": 1.0929, + "step": 790 + }, + { + "epoch": 0.09511212649552095, + "grad_norm": 2.5326778952656004, + "learning_rate": 3.955764557632634e-06, + "loss": 1.0019, + "step": 791 + }, + { + "epoch": 0.09523236938616005, + "grad_norm": 6.890377810103507, + "learning_rate": 3.955601482668309e-06, + "loss": 1.1879, + "step": 792 + }, + { + "epoch": 0.09535261227679913, + "grad_norm": 1.9118558659115137, + "learning_rate": 3.955438111040794e-06, + "loss": 1.1242, + "step": 793 + }, + { + "epoch": 0.09547285516743823, + "grad_norm": 2.1035496275977428, + "learning_rate": 3.955274442774873e-06, + "loss": 1.042, + "step": 794 + }, + { + "epoch": 0.09559309805807732, + "grad_norm": 2.115239559882265, + "learning_rate": 3.9551104778953725e-06, + "loss": 0.9445, + "step": 795 + }, + { + "epoch": 0.0957133409487164, + "grad_norm": 2.0809082667137297, + "learning_rate": 3.954946216427167e-06, + "loss": 1.0965, + "step": 796 + }, + { + "epoch": 0.0958335838393555, + "grad_norm": 0.9035528430125819, + "learning_rate": 3.954781658395176e-06, + "loss": 0.8542, + "step": 797 + }, + { + "epoch": 0.09595382672999458, + "grad_norm": 1.9793734258435287, + "learning_rate": 3.95461680382436e-06, + "loss": 1.1632, + "step": 798 + }, + { + "epoch": 0.09607406962063368, + "grad_norm": 6.19105689570366, + "learning_rate": 3.9544516527397295e-06, + "loss": 1.0954, + "step": 799 + }, + { + "epoch": 0.09619431251127276, + "grad_norm": 2.224103245417545, + "learning_rate": 3.954286205166338e-06, + "loss": 1.0448, + "step": 800 + }, + { + "epoch": 0.09631455540191186, + "grad_norm": 2.1061146723799515, + "learning_rate": 3.954120461129282e-06, + "loss": 1.0777, + "step": 801 + }, + { + "epoch": 0.09643479829255096, + "grad_norm": 2.7634745712753004, + "learning_rate": 3.953954420653706e-06, + "loss": 1.0774, + "step": 802 + }, + { + "epoch": 0.09655504118319004, + "grad_norm": 2.328260993825884, + "learning_rate": 3.953788083764798e-06, + "loss": 1.1176, + "step": 803 + }, + { + "epoch": 0.09667528407382914, + "grad_norm": 2.152638603825361, + "learning_rate": 3.953621450487792e-06, + "loss": 1.1618, + "step": 804 + }, + { + "epoch": 0.09679552696446822, + "grad_norm": 0.8371332831802448, + "learning_rate": 3.953454520847964e-06, + "loss": 0.8508, + "step": 805 + }, + { + "epoch": 0.09691576985510732, + "grad_norm": 2.092930875508841, + "learning_rate": 3.9532872948706395e-06, + "loss": 0.9727, + "step": 806 + }, + { + "epoch": 0.09703601274574641, + "grad_norm": 2.8732651161480627, + "learning_rate": 3.9531197725811845e-06, + "loss": 1.0645, + "step": 807 + }, + { + "epoch": 0.0971562556363855, + "grad_norm": 1.9161612079152521, + "learning_rate": 3.952951954005013e-06, + "loss": 1.1177, + "step": 808 + }, + { + "epoch": 0.0972764985270246, + "grad_norm": 1.798541150722859, + "learning_rate": 3.952783839167584e-06, + "loss": 1.0867, + "step": 809 + }, + { + "epoch": 0.09739674141766368, + "grad_norm": 2.5092764411088813, + "learning_rate": 3.952615428094398e-06, + "loss": 0.9843, + "step": 810 + }, + { + "epoch": 0.09751698430830277, + "grad_norm": 1.9356237371217022, + "learning_rate": 3.952446720811004e-06, + "loss": 0.9759, + "step": 811 + }, + { + "epoch": 0.09763722719894186, + "grad_norm": 0.8494960665507911, + "learning_rate": 3.952277717342995e-06, + "loss": 0.8741, + "step": 812 + }, + { + "epoch": 0.09775747008958095, + "grad_norm": 3.3639017522936774, + "learning_rate": 3.952108417716009e-06, + "loss": 1.0923, + "step": 813 + }, + { + "epoch": 0.09787771298022005, + "grad_norm": 4.187804253755592, + "learning_rate": 3.951938821955727e-06, + "loss": 1.0892, + "step": 814 + }, + { + "epoch": 0.09799795587085913, + "grad_norm": 2.2263782090329984, + "learning_rate": 3.9517689300878786e-06, + "loss": 1.0044, + "step": 815 + }, + { + "epoch": 0.09811819876149823, + "grad_norm": 2.0930123239547154, + "learning_rate": 3.951598742138236e-06, + "loss": 1.026, + "step": 816 + }, + { + "epoch": 0.09823844165213731, + "grad_norm": 2.587196804103163, + "learning_rate": 3.951428258132615e-06, + "loss": 1.038, + "step": 817 + }, + { + "epoch": 0.09835868454277641, + "grad_norm": 2.327362550145408, + "learning_rate": 3.951257478096879e-06, + "loss": 1.0854, + "step": 818 + }, + { + "epoch": 0.0984789274334155, + "grad_norm": 3.1128877513143185, + "learning_rate": 3.951086402056936e-06, + "loss": 0.9272, + "step": 819 + }, + { + "epoch": 0.09859917032405459, + "grad_norm": 1.9483417380385257, + "learning_rate": 3.950915030038735e-06, + "loss": 1.0752, + "step": 820 + }, + { + "epoch": 0.09871941321469369, + "grad_norm": 2.724909327687109, + "learning_rate": 3.9507433620682765e-06, + "loss": 1.0745, + "step": 821 + }, + { + "epoch": 0.09883965610533277, + "grad_norm": 2.0461214731500528, + "learning_rate": 3.9505713981716e-06, + "loss": 1.1217, + "step": 822 + }, + { + "epoch": 0.09895989899597187, + "grad_norm": 2.2340855977898326, + "learning_rate": 3.950399138374795e-06, + "loss": 1.0482, + "step": 823 + }, + { + "epoch": 0.09908014188661095, + "grad_norm": 2.01224915262044, + "learning_rate": 3.95022658270399e-06, + "loss": 0.9853, + "step": 824 + }, + { + "epoch": 0.09920038477725004, + "grad_norm": 3.574654040903259, + "learning_rate": 3.9500537311853635e-06, + "loss": 1.0181, + "step": 825 + }, + { + "epoch": 0.09932062766788914, + "grad_norm": 5.909403287193501, + "learning_rate": 3.949880583845136e-06, + "loss": 1.0782, + "step": 826 + }, + { + "epoch": 0.09944087055852822, + "grad_norm": 3.704247928124236, + "learning_rate": 3.949707140709575e-06, + "loss": 1.0513, + "step": 827 + }, + { + "epoch": 0.09956111344916732, + "grad_norm": 5.430461669179674, + "learning_rate": 3.949533401804991e-06, + "loss": 1.0756, + "step": 828 + }, + { + "epoch": 0.0996813563398064, + "grad_norm": 1.9814094551493082, + "learning_rate": 3.949359367157739e-06, + "loss": 1.1495, + "step": 829 + }, + { + "epoch": 0.0998015992304455, + "grad_norm": 2.221807388156849, + "learning_rate": 3.949185036794222e-06, + "loss": 0.9938, + "step": 830 + }, + { + "epoch": 0.0999218421210846, + "grad_norm": 1.859904814817488, + "learning_rate": 3.949010410740884e-06, + "loss": 1.0288, + "step": 831 + }, + { + "epoch": 0.10004208501172368, + "grad_norm": 1.8630081712397177, + "learning_rate": 3.948835489024216e-06, + "loss": 1.1069, + "step": 832 + }, + { + "epoch": 0.10016232790236278, + "grad_norm": 2.418072576558477, + "learning_rate": 3.948660271670755e-06, + "loss": 1.1408, + "step": 833 + }, + { + "epoch": 0.10028257079300186, + "grad_norm": 2.2971472546380145, + "learning_rate": 3.948484758707079e-06, + "loss": 1.0823, + "step": 834 + }, + { + "epoch": 0.10040281368364096, + "grad_norm": 2.659952970963821, + "learning_rate": 3.948308950159815e-06, + "loss": 1.0739, + "step": 835 + }, + { + "epoch": 0.10052305657428004, + "grad_norm": 2.691280456172194, + "learning_rate": 3.9481328460556326e-06, + "loss": 0.9968, + "step": 836 + }, + { + "epoch": 0.10064329946491914, + "grad_norm": 4.150446231245145, + "learning_rate": 3.9479564464212455e-06, + "loss": 1.1343, + "step": 837 + }, + { + "epoch": 0.10076354235555823, + "grad_norm": 3.9997019656730655, + "learning_rate": 3.947779751283414e-06, + "loss": 1.0037, + "step": 838 + }, + { + "epoch": 0.10088378524619732, + "grad_norm": 2.1241726106022005, + "learning_rate": 3.947602760668944e-06, + "loss": 1.0014, + "step": 839 + }, + { + "epoch": 0.10100402813683641, + "grad_norm": 2.1742336720904794, + "learning_rate": 3.947425474604684e-06, + "loss": 0.9564, + "step": 840 + }, + { + "epoch": 0.1011242710274755, + "grad_norm": 2.5822744557949817, + "learning_rate": 3.947247893117528e-06, + "loss": 1.164, + "step": 841 + }, + { + "epoch": 0.10124451391811459, + "grad_norm": 3.2983061228961863, + "learning_rate": 3.947070016234413e-06, + "loss": 0.9361, + "step": 842 + }, + { + "epoch": 0.10136475680875369, + "grad_norm": 3.5159915987246393, + "learning_rate": 3.946891843982326e-06, + "loss": 0.9878, + "step": 843 + }, + { + "epoch": 0.10148499969939277, + "grad_norm": 2.3791872556990263, + "learning_rate": 3.9467133763882935e-06, + "loss": 0.9818, + "step": 844 + }, + { + "epoch": 0.10160524259003187, + "grad_norm": 1.9364379925520527, + "learning_rate": 3.9465346134793905e-06, + "loss": 1.1012, + "step": 845 + }, + { + "epoch": 0.10172548548067095, + "grad_norm": 2.4752433464562684, + "learning_rate": 3.9463555552827335e-06, + "loss": 1.0331, + "step": 846 + }, + { + "epoch": 0.10184572837131005, + "grad_norm": 3.168457896596526, + "learning_rate": 3.946176201825487e-06, + "loss": 1.0982, + "step": 847 + }, + { + "epoch": 0.10196597126194913, + "grad_norm": 2.5470563203824996, + "learning_rate": 3.9459965531348575e-06, + "loss": 1.0729, + "step": 848 + }, + { + "epoch": 0.10208621415258823, + "grad_norm": 4.431149518981537, + "learning_rate": 3.945816609238098e-06, + "loss": 1.0935, + "step": 849 + }, + { + "epoch": 0.10220645704322733, + "grad_norm": 2.3573245808293612, + "learning_rate": 3.945636370162507e-06, + "loss": 1.0899, + "step": 850 + }, + { + "epoch": 0.10232669993386641, + "grad_norm": 1.9905926472716176, + "learning_rate": 3.945455835935425e-06, + "loss": 1.0296, + "step": 851 + }, + { + "epoch": 0.1024469428245055, + "grad_norm": 2.6210754666778033, + "learning_rate": 3.94527500658424e-06, + "loss": 0.9932, + "step": 852 + }, + { + "epoch": 0.10256718571514459, + "grad_norm": 2.6215717999689763, + "learning_rate": 3.945093882136382e-06, + "loss": 1.0539, + "step": 853 + }, + { + "epoch": 0.10268742860578368, + "grad_norm": 2.3279607318043922, + "learning_rate": 3.944912462619329e-06, + "loss": 1.0842, + "step": 854 + }, + { + "epoch": 0.10280767149642277, + "grad_norm": 2.315723749647219, + "learning_rate": 3.9447307480606025e-06, + "loss": 1.0465, + "step": 855 + }, + { + "epoch": 0.10292791438706186, + "grad_norm": 3.5847365418994324, + "learning_rate": 3.944548738487767e-06, + "loss": 1.1411, + "step": 856 + }, + { + "epoch": 0.10304815727770096, + "grad_norm": 3.0010091355924224, + "learning_rate": 3.944366433928434e-06, + "loss": 1.1472, + "step": 857 + }, + { + "epoch": 0.10316840016834004, + "grad_norm": 3.080520240146325, + "learning_rate": 3.9441838344102594e-06, + "loss": 1.0752, + "step": 858 + }, + { + "epoch": 0.10328864305897914, + "grad_norm": 2.5773005236481015, + "learning_rate": 3.944000939960943e-06, + "loss": 0.9148, + "step": 859 + }, + { + "epoch": 0.10340888594961822, + "grad_norm": 2.9066488340979864, + "learning_rate": 3.943817750608229e-06, + "loss": 1.0351, + "step": 860 + }, + { + "epoch": 0.10352912884025732, + "grad_norm": 2.7881908604108094, + "learning_rate": 3.943634266379908e-06, + "loss": 1.0605, + "step": 861 + }, + { + "epoch": 0.10364937173089642, + "grad_norm": 1.890392762099498, + "learning_rate": 3.943450487303815e-06, + "loss": 1.0904, + "step": 862 + }, + { + "epoch": 0.1037696146215355, + "grad_norm": 2.7617500012044593, + "learning_rate": 3.943266413407827e-06, + "loss": 1.091, + "step": 863 + }, + { + "epoch": 0.1038898575121746, + "grad_norm": 4.31344483570939, + "learning_rate": 3.94308204471987e-06, + "loss": 1.0892, + "step": 864 + }, + { + "epoch": 0.10401010040281368, + "grad_norm": 2.421706359898532, + "learning_rate": 3.942897381267912e-06, + "loss": 0.9823, + "step": 865 + }, + { + "epoch": 0.10413034329345278, + "grad_norm": 2.602768169002196, + "learning_rate": 3.942712423079965e-06, + "loss": 0.901, + "step": 866 + }, + { + "epoch": 0.10425058618409186, + "grad_norm": 4.105449237287122, + "learning_rate": 3.942527170184088e-06, + "loss": 1.1364, + "step": 867 + }, + { + "epoch": 0.10437082907473096, + "grad_norm": 2.4576111611015974, + "learning_rate": 3.942341622608385e-06, + "loss": 1.0175, + "step": 868 + }, + { + "epoch": 0.10449107196537005, + "grad_norm": 1.8845952248505622, + "learning_rate": 3.942155780381001e-06, + "loss": 1.0159, + "step": 869 + }, + { + "epoch": 0.10461131485600914, + "grad_norm": 2.5878458501433204, + "learning_rate": 3.94196964353013e-06, + "loss": 1.0021, + "step": 870 + }, + { + "epoch": 0.10473155774664823, + "grad_norm": 1.9840381366425337, + "learning_rate": 3.941783212084008e-06, + "loss": 1.0453, + "step": 871 + }, + { + "epoch": 0.10485180063728732, + "grad_norm": 3.277790680193606, + "learning_rate": 3.941596486070916e-06, + "loss": 1.0215, + "step": 872 + }, + { + "epoch": 0.10497204352792641, + "grad_norm": 2.51044456709135, + "learning_rate": 3.941409465519182e-06, + "loss": 0.8279, + "step": 873 + }, + { + "epoch": 0.10509228641856551, + "grad_norm": 1.8379418652776753, + "learning_rate": 3.941222150457176e-06, + "loss": 1.0946, + "step": 874 + }, + { + "epoch": 0.10521252930920459, + "grad_norm": 3.0140867114639605, + "learning_rate": 3.941034540913311e-06, + "loss": 0.9545, + "step": 875 + }, + { + "epoch": 0.10533277219984369, + "grad_norm": 1.9246057478210696, + "learning_rate": 3.940846636916051e-06, + "loss": 1.0626, + "step": 876 + }, + { + "epoch": 0.10545301509048277, + "grad_norm": 1.9969628638576293, + "learning_rate": 3.940658438493899e-06, + "loss": 1.1045, + "step": 877 + }, + { + "epoch": 0.10557325798112187, + "grad_norm": 3.0979991977954118, + "learning_rate": 3.940469945675405e-06, + "loss": 0.9991, + "step": 878 + }, + { + "epoch": 0.10569350087176095, + "grad_norm": 2.0032318943680796, + "learning_rate": 3.940281158489163e-06, + "loss": 1.153, + "step": 879 + }, + { + "epoch": 0.10581374376240005, + "grad_norm": 2.6087720711198927, + "learning_rate": 3.940092076963812e-06, + "loss": 1.0668, + "step": 880 + }, + { + "epoch": 0.10593398665303914, + "grad_norm": 2.3612847083606816, + "learning_rate": 3.9399027011280355e-06, + "loss": 1.0287, + "step": 881 + }, + { + "epoch": 0.10605422954367823, + "grad_norm": 3.7071584633483754, + "learning_rate": 3.939713031010561e-06, + "loss": 1.0114, + "step": 882 + }, + { + "epoch": 0.10617447243431732, + "grad_norm": 2.014337409340708, + "learning_rate": 3.939523066640163e-06, + "loss": 1.016, + "step": 883 + }, + { + "epoch": 0.10629471532495641, + "grad_norm": 1.8740257911210143, + "learning_rate": 3.939332808045657e-06, + "loss": 1.0479, + "step": 884 + }, + { + "epoch": 0.1064149582155955, + "grad_norm": 2.1505068159458802, + "learning_rate": 3.939142255255906e-06, + "loss": 1.0857, + "step": 885 + }, + { + "epoch": 0.1065352011062346, + "grad_norm": 2.394559228431825, + "learning_rate": 3.938951408299817e-06, + "loss": 1.1082, + "step": 886 + }, + { + "epoch": 0.10665544399687368, + "grad_norm": 0.8027938486607697, + "learning_rate": 3.938760267206342e-06, + "loss": 0.7828, + "step": 887 + }, + { + "epoch": 0.10677568688751278, + "grad_norm": 2.7331266332796367, + "learning_rate": 3.938568832004475e-06, + "loss": 1.0228, + "step": 888 + }, + { + "epoch": 0.10689592977815186, + "grad_norm": 2.1427770781318767, + "learning_rate": 3.938377102723257e-06, + "loss": 0.9925, + "step": 889 + }, + { + "epoch": 0.10701617266879096, + "grad_norm": 2.7303295618471384, + "learning_rate": 3.938185079391774e-06, + "loss": 1.0704, + "step": 890 + }, + { + "epoch": 0.10713641555943004, + "grad_norm": 3.271876434063529, + "learning_rate": 3.937992762039157e-06, + "loss": 1.2983, + "step": 891 + }, + { + "epoch": 0.10725665845006914, + "grad_norm": 1.8254233718373924, + "learning_rate": 3.937800150694577e-06, + "loss": 1.0463, + "step": 892 + }, + { + "epoch": 0.10737690134070824, + "grad_norm": 2.3341369834391674, + "learning_rate": 3.937607245387255e-06, + "loss": 0.9978, + "step": 893 + }, + { + "epoch": 0.10749714423134732, + "grad_norm": 2.097063772968173, + "learning_rate": 3.937414046146455e-06, + "loss": 0.9642, + "step": 894 + }, + { + "epoch": 0.10761738712198642, + "grad_norm": 2.137264983882, + "learning_rate": 3.9372205530014845e-06, + "loss": 0.994, + "step": 895 + }, + { + "epoch": 0.1077376300126255, + "grad_norm": 2.5820660610420494, + "learning_rate": 3.937026765981696e-06, + "loss": 0.9528, + "step": 896 + }, + { + "epoch": 0.1078578729032646, + "grad_norm": 1.8187688085887823, + "learning_rate": 3.936832685116488e-06, + "loss": 1.0286, + "step": 897 + }, + { + "epoch": 0.10797811579390369, + "grad_norm": 2.5634023426755643, + "learning_rate": 3.936638310435301e-06, + "loss": 1.1421, + "step": 898 + }, + { + "epoch": 0.10809835868454278, + "grad_norm": 2.5539026017044817, + "learning_rate": 3.936443641967623e-06, + "loss": 1.0521, + "step": 899 + }, + { + "epoch": 0.10821860157518187, + "grad_norm": 2.204566213622667, + "learning_rate": 3.936248679742983e-06, + "loss": 1.067, + "step": 900 + }, + { + "epoch": 0.10833884446582095, + "grad_norm": 1.0690365167452545, + "learning_rate": 3.936053423790959e-06, + "loss": 0.941, + "step": 901 + }, + { + "epoch": 0.10845908735646005, + "grad_norm": 1.6948163409587154, + "learning_rate": 3.935857874141168e-06, + "loss": 1.0134, + "step": 902 + }, + { + "epoch": 0.10857933024709913, + "grad_norm": 2.454420488721385, + "learning_rate": 3.935662030823279e-06, + "loss": 1.0773, + "step": 903 + }, + { + "epoch": 0.10869957313773823, + "grad_norm": 3.094347501142156, + "learning_rate": 3.935465893866998e-06, + "loss": 0.96, + "step": 904 + }, + { + "epoch": 0.10881981602837733, + "grad_norm": 2.694398586848652, + "learning_rate": 3.935269463302079e-06, + "loss": 1.0411, + "step": 905 + }, + { + "epoch": 0.10894005891901641, + "grad_norm": 1.942546547978826, + "learning_rate": 3.935072739158322e-06, + "loss": 1.0084, + "step": 906 + }, + { + "epoch": 0.10906030180965551, + "grad_norm": 1.741644257153378, + "learning_rate": 3.934875721465569e-06, + "loss": 1.0376, + "step": 907 + }, + { + "epoch": 0.10918054470029459, + "grad_norm": 3.462016675797645, + "learning_rate": 3.9346784102537076e-06, + "loss": 0.9526, + "step": 908 + }, + { + "epoch": 0.10930078759093369, + "grad_norm": 2.008806034771215, + "learning_rate": 3.934480805552669e-06, + "loss": 1.0222, + "step": 909 + }, + { + "epoch": 0.10942103048157277, + "grad_norm": 3.747625107563239, + "learning_rate": 3.93428290739243e-06, + "loss": 1.1202, + "step": 910 + }, + { + "epoch": 0.10954127337221187, + "grad_norm": 5.851219258854136, + "learning_rate": 3.9340847158030125e-06, + "loss": 1.0387, + "step": 911 + }, + { + "epoch": 0.10966151626285096, + "grad_norm": 29.165594404583224, + "learning_rate": 3.9338862308144814e-06, + "loss": 0.9962, + "step": 912 + }, + { + "epoch": 0.10978175915349005, + "grad_norm": 1.9018989058542517, + "learning_rate": 3.933687452456946e-06, + "loss": 1.085, + "step": 913 + }, + { + "epoch": 0.10990200204412914, + "grad_norm": 3.6119512966247833, + "learning_rate": 3.933488380760562e-06, + "loss": 1.1017, + "step": 914 + }, + { + "epoch": 0.11002224493476823, + "grad_norm": 3.4307738649521866, + "learning_rate": 3.9332890157555286e-06, + "loss": 1.1105, + "step": 915 + }, + { + "epoch": 0.11014248782540732, + "grad_norm": 2.547113231914244, + "learning_rate": 3.933089357472088e-06, + "loss": 0.9994, + "step": 916 + }, + { + "epoch": 0.11026273071604642, + "grad_norm": 2.3661434275733653, + "learning_rate": 3.932889405940529e-06, + "loss": 1.1024, + "step": 917 + }, + { + "epoch": 0.1103829736066855, + "grad_norm": 3.174704957004017, + "learning_rate": 3.932689161191184e-06, + "loss": 1.0395, + "step": 918 + }, + { + "epoch": 0.1105032164973246, + "grad_norm": 2.1858309236494544, + "learning_rate": 3.93248862325443e-06, + "loss": 1.1192, + "step": 919 + }, + { + "epoch": 0.11062345938796368, + "grad_norm": 0.938099002211052, + "learning_rate": 3.932287792160688e-06, + "loss": 0.8839, + "step": 920 + }, + { + "epoch": 0.11074370227860278, + "grad_norm": 2.5940953220156286, + "learning_rate": 3.932086667940424e-06, + "loss": 1.0407, + "step": 921 + }, + { + "epoch": 0.11086394516924186, + "grad_norm": 1.9282647575289482, + "learning_rate": 3.93188525062415e-06, + "loss": 1.055, + "step": 922 + }, + { + "epoch": 0.11098418805988096, + "grad_norm": 1.9397975465984978, + "learning_rate": 3.931683540242418e-06, + "loss": 1.0992, + "step": 923 + }, + { + "epoch": 0.11110443095052006, + "grad_norm": 3.887862975417858, + "learning_rate": 3.9314815368258295e-06, + "loss": 1.1497, + "step": 924 + }, + { + "epoch": 0.11122467384115914, + "grad_norm": 1.8476410234077925, + "learning_rate": 3.9312792404050275e-06, + "loss": 1.0279, + "step": 925 + }, + { + "epoch": 0.11134491673179824, + "grad_norm": 1.8964681575564966, + "learning_rate": 3.9310766510107e-06, + "loss": 1.015, + "step": 926 + }, + { + "epoch": 0.11146515962243732, + "grad_norm": 3.82786291354109, + "learning_rate": 3.9308737686735806e-06, + "loss": 1.1612, + "step": 927 + }, + { + "epoch": 0.11158540251307641, + "grad_norm": 2.1676203267314538, + "learning_rate": 3.9306705934244455e-06, + "loss": 1.0646, + "step": 928 + }, + { + "epoch": 0.11170564540371551, + "grad_norm": 1.843763319064379, + "learning_rate": 3.930467125294116e-06, + "loss": 1.1194, + "step": 929 + }, + { + "epoch": 0.1118258882943546, + "grad_norm": 0.9442744269550757, + "learning_rate": 3.930263364313458e-06, + "loss": 0.8433, + "step": 930 + }, + { + "epoch": 0.11194613118499369, + "grad_norm": 2.350999327483866, + "learning_rate": 3.930059310513384e-06, + "loss": 1.0706, + "step": 931 + }, + { + "epoch": 0.11206637407563277, + "grad_norm": 1.8445730150495863, + "learning_rate": 3.929854963924846e-06, + "loss": 1.0801, + "step": 932 + }, + { + "epoch": 0.11218661696627187, + "grad_norm": 2.0547290406179806, + "learning_rate": 3.929650324578845e-06, + "loss": 1.014, + "step": 933 + }, + { + "epoch": 0.11230685985691095, + "grad_norm": 4.1640304363798375, + "learning_rate": 3.929445392506423e-06, + "loss": 1.0569, + "step": 934 + }, + { + "epoch": 0.11242710274755005, + "grad_norm": 2.067469072952422, + "learning_rate": 3.92924016773867e-06, + "loss": 0.9958, + "step": 935 + }, + { + "epoch": 0.11254734563818915, + "grad_norm": 2.3861282042588208, + "learning_rate": 3.9290346503067175e-06, + "loss": 0.973, + "step": 936 + }, + { + "epoch": 0.11266758852882823, + "grad_norm": 1.9692333778492448, + "learning_rate": 3.9288288402417415e-06, + "loss": 1.0299, + "step": 937 + }, + { + "epoch": 0.11278783141946733, + "grad_norm": 2.6649907528142474, + "learning_rate": 3.928622737574964e-06, + "loss": 0.945, + "step": 938 + }, + { + "epoch": 0.11290807431010641, + "grad_norm": 3.704774384624089, + "learning_rate": 3.928416342337652e-06, + "loss": 1.1511, + "step": 939 + }, + { + "epoch": 0.1130283172007455, + "grad_norm": 2.087583999314307, + "learning_rate": 3.928209654561113e-06, + "loss": 1.0641, + "step": 940 + }, + { + "epoch": 0.1131485600913846, + "grad_norm": 2.4609181236836406, + "learning_rate": 3.928002674276703e-06, + "loss": 1.0532, + "step": 941 + }, + { + "epoch": 0.11326880298202369, + "grad_norm": 2.3594439161732357, + "learning_rate": 3.92779540151582e-06, + "loss": 0.9944, + "step": 942 + }, + { + "epoch": 0.11338904587266278, + "grad_norm": 2.231370199262329, + "learning_rate": 3.927587836309907e-06, + "loss": 1.094, + "step": 943 + }, + { + "epoch": 0.11350928876330187, + "grad_norm": 4.4178649188589025, + "learning_rate": 3.927379978690452e-06, + "loss": 1.022, + "step": 944 + }, + { + "epoch": 0.11362953165394096, + "grad_norm": 2.3240470245777582, + "learning_rate": 3.927171828688987e-06, + "loss": 1.1101, + "step": 945 + }, + { + "epoch": 0.11374977454458005, + "grad_norm": 2.242166552296257, + "learning_rate": 3.926963386337088e-06, + "loss": 1.0569, + "step": 946 + }, + { + "epoch": 0.11387001743521914, + "grad_norm": 2.1722583020859756, + "learning_rate": 3.926754651666375e-06, + "loss": 0.942, + "step": 947 + }, + { + "epoch": 0.11399026032585824, + "grad_norm": 3.077903933259693, + "learning_rate": 3.926545624708513e-06, + "loss": 1.0193, + "step": 948 + }, + { + "epoch": 0.11411050321649732, + "grad_norm": 2.0932225943793847, + "learning_rate": 3.926336305495213e-06, + "loss": 1.0953, + "step": 949 + }, + { + "epoch": 0.11423074610713642, + "grad_norm": 4.076597197952659, + "learning_rate": 3.926126694058226e-06, + "loss": 1.1244, + "step": 950 + }, + { + "epoch": 0.1143509889977755, + "grad_norm": 1.8792426111589369, + "learning_rate": 3.92591679042935e-06, + "loss": 1.0601, + "step": 951 + }, + { + "epoch": 0.1144712318884146, + "grad_norm": 2.6192047088543435, + "learning_rate": 3.92570659464043e-06, + "loss": 1.0593, + "step": 952 + }, + { + "epoch": 0.1145914747790537, + "grad_norm": 2.1272692064554026, + "learning_rate": 3.925496106723349e-06, + "loss": 1.0349, + "step": 953 + }, + { + "epoch": 0.11471171766969278, + "grad_norm": 1.9914896023091224, + "learning_rate": 3.9252853267100405e-06, + "loss": 1.0748, + "step": 954 + }, + { + "epoch": 0.11483196056033187, + "grad_norm": 1.8247773909994833, + "learning_rate": 3.9250742546324786e-06, + "loss": 1.0781, + "step": 955 + }, + { + "epoch": 0.11495220345097096, + "grad_norm": 2.082207248255735, + "learning_rate": 3.924862890522683e-06, + "loss": 1.1101, + "step": 956 + }, + { + "epoch": 0.11507244634161005, + "grad_norm": 2.44509486975846, + "learning_rate": 3.9246512344127174e-06, + "loss": 1.104, + "step": 957 + }, + { + "epoch": 0.11519268923224914, + "grad_norm": 1.8727980081426236, + "learning_rate": 3.9244392863346895e-06, + "loss": 1.0574, + "step": 958 + }, + { + "epoch": 0.11531293212288823, + "grad_norm": 2.1800490869017763, + "learning_rate": 3.9242270463207524e-06, + "loss": 1.1657, + "step": 959 + }, + { + "epoch": 0.11543317501352733, + "grad_norm": 2.578736481254375, + "learning_rate": 3.924014514403102e-06, + "loss": 1.0927, + "step": 960 + }, + { + "epoch": 0.11555341790416641, + "grad_norm": 2.232886505502207, + "learning_rate": 3.92380169061398e-06, + "loss": 1.1486, + "step": 961 + }, + { + "epoch": 0.11567366079480551, + "grad_norm": 2.5653155606038167, + "learning_rate": 3.9235885749856705e-06, + "loss": 1.077, + "step": 962 + }, + { + "epoch": 0.1157939036854446, + "grad_norm": 2.7128134621068196, + "learning_rate": 3.9233751675505035e-06, + "loss": 1.0655, + "step": 963 + }, + { + "epoch": 0.11591414657608369, + "grad_norm": 2.2483746698122338, + "learning_rate": 3.923161468340853e-06, + "loss": 1.088, + "step": 964 + }, + { + "epoch": 0.11603438946672277, + "grad_norm": 2.186020705171271, + "learning_rate": 3.9229474773891374e-06, + "loss": 1.0558, + "step": 965 + }, + { + "epoch": 0.11615463235736187, + "grad_norm": 2.262834395676869, + "learning_rate": 3.922733194727818e-06, + "loss": 1.0767, + "step": 966 + }, + { + "epoch": 0.11627487524800097, + "grad_norm": 2.167247095368129, + "learning_rate": 3.922518620389402e-06, + "loss": 1.1128, + "step": 967 + }, + { + "epoch": 0.11639511813864005, + "grad_norm": 1.8504438280280913, + "learning_rate": 3.922303754406439e-06, + "loss": 1.1338, + "step": 968 + }, + { + "epoch": 0.11651536102927915, + "grad_norm": 2.5363963052318153, + "learning_rate": 3.922088596811526e-06, + "loss": 1.0297, + "step": 969 + }, + { + "epoch": 0.11663560391991823, + "grad_norm": 2.698655153827963, + "learning_rate": 3.9218731476373e-06, + "loss": 1.1104, + "step": 970 + }, + { + "epoch": 0.11675584681055733, + "grad_norm": 3.00555842766318, + "learning_rate": 3.9216574069164455e-06, + "loss": 1.0874, + "step": 971 + }, + { + "epoch": 0.11687608970119642, + "grad_norm": 1.557471155478163, + "learning_rate": 3.921441374681691e-06, + "loss": 1.0424, + "step": 972 + }, + { + "epoch": 0.1169963325918355, + "grad_norm": 2.455167855236724, + "learning_rate": 3.921225050965808e-06, + "loss": 0.89, + "step": 973 + }, + { + "epoch": 0.1171165754824746, + "grad_norm": 2.0948686031387513, + "learning_rate": 3.921008435801612e-06, + "loss": 0.9876, + "step": 974 + }, + { + "epoch": 0.11723681837311369, + "grad_norm": 2.5026037009110085, + "learning_rate": 3.920791529221963e-06, + "loss": 0.9984, + "step": 975 + }, + { + "epoch": 0.11735706126375278, + "grad_norm": 5.589512248042803, + "learning_rate": 3.920574331259768e-06, + "loss": 1.0033, + "step": 976 + }, + { + "epoch": 0.11747730415439187, + "grad_norm": 2.507892833147966, + "learning_rate": 3.9203568419479716e-06, + "loss": 1.0339, + "step": 977 + }, + { + "epoch": 0.11759754704503096, + "grad_norm": 1.9560065413523628, + "learning_rate": 3.92013906131957e-06, + "loss": 0.9901, + "step": 978 + }, + { + "epoch": 0.11771778993567006, + "grad_norm": 1.5644003946130693, + "learning_rate": 3.9199209894076e-06, + "loss": 1.0633, + "step": 979 + }, + { + "epoch": 0.11783803282630914, + "grad_norm": 2.0753980829243823, + "learning_rate": 3.919702626245142e-06, + "loss": 1.1378, + "step": 980 + }, + { + "epoch": 0.11795827571694824, + "grad_norm": 2.221337542335641, + "learning_rate": 3.919483971865322e-06, + "loss": 0.8971, + "step": 981 + }, + { + "epoch": 0.11807851860758732, + "grad_norm": 2.2394751116014167, + "learning_rate": 3.91926502630131e-06, + "loss": 1.1178, + "step": 982 + }, + { + "epoch": 0.11819876149822642, + "grad_norm": 2.133394378543699, + "learning_rate": 3.91904578958632e-06, + "loss": 0.9654, + "step": 983 + }, + { + "epoch": 0.11831900438886551, + "grad_norm": 2.136936992539323, + "learning_rate": 3.918826261753608e-06, + "loss": 1.0817, + "step": 984 + }, + { + "epoch": 0.1184392472795046, + "grad_norm": 3.0341449559989173, + "learning_rate": 3.918606442836478e-06, + "loss": 0.9523, + "step": 985 + }, + { + "epoch": 0.1185594901701437, + "grad_norm": 2.04146308921989, + "learning_rate": 3.918386332868277e-06, + "loss": 1.0103, + "step": 986 + }, + { + "epoch": 0.11867973306078278, + "grad_norm": 2.169419911281549, + "learning_rate": 3.918165931882394e-06, + "loss": 1.1808, + "step": 987 + }, + { + "epoch": 0.11879997595142187, + "grad_norm": 4.8250619993031325, + "learning_rate": 3.917945239912264e-06, + "loss": 0.9931, + "step": 988 + }, + { + "epoch": 0.11892021884206096, + "grad_norm": 2.246404954663543, + "learning_rate": 3.917724256991367e-06, + "loss": 0.9951, + "step": 989 + }, + { + "epoch": 0.11904046173270005, + "grad_norm": 2.7914282758662132, + "learning_rate": 3.9175029831532245e-06, + "loss": 1.0535, + "step": 990 + }, + { + "epoch": 0.11916070462333915, + "grad_norm": 2.59815577243093, + "learning_rate": 3.917281418431404e-06, + "loss": 1.1282, + "step": 991 + }, + { + "epoch": 0.11928094751397823, + "grad_norm": 2.0370455884593426, + "learning_rate": 3.917059562859516e-06, + "loss": 1.0079, + "step": 992 + }, + { + "epoch": 0.11940119040461733, + "grad_norm": 2.6929804789235443, + "learning_rate": 3.916837416471218e-06, + "loss": 1.1282, + "step": 993 + }, + { + "epoch": 0.11952143329525641, + "grad_norm": 4.281176044354391, + "learning_rate": 3.916614979300207e-06, + "loss": 0.9613, + "step": 994 + }, + { + "epoch": 0.11964167618589551, + "grad_norm": 1.6285503922683728, + "learning_rate": 3.9163922513802274e-06, + "loss": 1.0265, + "step": 995 + }, + { + "epoch": 0.1197619190765346, + "grad_norm": 3.0717271646995323, + "learning_rate": 3.916169232745067e-06, + "loss": 1.066, + "step": 996 + }, + { + "epoch": 0.11988216196717369, + "grad_norm": 2.5141309009301245, + "learning_rate": 3.915945923428559e-06, + "loss": 1.1606, + "step": 997 + }, + { + "epoch": 0.12000240485781279, + "grad_norm": 2.231131380064254, + "learning_rate": 3.915722323464577e-06, + "loss": 1.071, + "step": 998 + }, + { + "epoch": 0.12012264774845187, + "grad_norm": 2.49428381691787, + "learning_rate": 3.91549843288704e-06, + "loss": 0.9458, + "step": 999 + }, + { + "epoch": 0.12024289063909097, + "grad_norm": 1.950764625976518, + "learning_rate": 3.915274251729916e-06, + "loss": 1.0311, + "step": 1000 + }, + { + "epoch": 0.12036313352973005, + "grad_norm": 2.1553139244186608, + "learning_rate": 3.91504978002721e-06, + "loss": 1.1399, + "step": 1001 + }, + { + "epoch": 0.12048337642036915, + "grad_norm": 4.384228809010298, + "learning_rate": 3.914825017812974e-06, + "loss": 1.002, + "step": 1002 + }, + { + "epoch": 0.12060361931100824, + "grad_norm": 2.5513751753695386, + "learning_rate": 3.9145999651213065e-06, + "loss": 0.9624, + "step": 1003 + }, + { + "epoch": 0.12072386220164733, + "grad_norm": 2.919114874633431, + "learning_rate": 3.9143746219863465e-06, + "loss": 1.1257, + "step": 1004 + }, + { + "epoch": 0.12084410509228642, + "grad_norm": 0.9709054287075727, + "learning_rate": 3.914148988442278e-06, + "loss": 0.8859, + "step": 1005 + }, + { + "epoch": 0.1209643479829255, + "grad_norm": 6.792209751749984, + "learning_rate": 3.91392306452333e-06, + "loss": 1.1961, + "step": 1006 + }, + { + "epoch": 0.1210845908735646, + "grad_norm": 3.6242918440924132, + "learning_rate": 3.913696850263774e-06, + "loss": 0.9082, + "step": 1007 + }, + { + "epoch": 0.1212048337642037, + "grad_norm": 2.296084001798856, + "learning_rate": 3.913470345697929e-06, + "loss": 1.0265, + "step": 1008 + }, + { + "epoch": 0.12132507665484278, + "grad_norm": 3.338612111856521, + "learning_rate": 3.913243550860153e-06, + "loss": 1.094, + "step": 1009 + }, + { + "epoch": 0.12144531954548188, + "grad_norm": 2.3936150495372255, + "learning_rate": 3.913016465784852e-06, + "loss": 1.0014, + "step": 1010 + }, + { + "epoch": 0.12156556243612096, + "grad_norm": 2.903231293833753, + "learning_rate": 3.912789090506474e-06, + "loss": 0.9544, + "step": 1011 + }, + { + "epoch": 0.12168580532676006, + "grad_norm": 2.3104373909170124, + "learning_rate": 3.9125614250595114e-06, + "loss": 0.9562, + "step": 1012 + }, + { + "epoch": 0.12180604821739914, + "grad_norm": 2.8079734828272915, + "learning_rate": 3.912333469478502e-06, + "loss": 1.1322, + "step": 1013 + }, + { + "epoch": 0.12192629110803824, + "grad_norm": 3.9079660537244933, + "learning_rate": 3.912105223798025e-06, + "loss": 1.019, + "step": 1014 + }, + { + "epoch": 0.12204653399867733, + "grad_norm": 1.0003061421986883, + "learning_rate": 3.9118766880527065e-06, + "loss": 0.9152, + "step": 1015 + }, + { + "epoch": 0.12216677688931642, + "grad_norm": 1.9244992087787003, + "learning_rate": 3.9116478622772145e-06, + "loss": 0.9761, + "step": 1016 + }, + { + "epoch": 0.12228701977995551, + "grad_norm": 1.916201645651014, + "learning_rate": 3.911418746506261e-06, + "loss": 1.1221, + "step": 1017 + }, + { + "epoch": 0.1224072626705946, + "grad_norm": 1.9631134362230864, + "learning_rate": 3.911189340774604e-06, + "loss": 1.0229, + "step": 1018 + }, + { + "epoch": 0.1225275055612337, + "grad_norm": 3.17707826374254, + "learning_rate": 3.910959645117043e-06, + "loss": 1.0308, + "step": 1019 + }, + { + "epoch": 0.12264774845187278, + "grad_norm": 0.8225069756487448, + "learning_rate": 3.910729659568423e-06, + "loss": 0.805, + "step": 1020 + }, + { + "epoch": 0.12276799134251187, + "grad_norm": 1.978180234598926, + "learning_rate": 3.9104993841636344e-06, + "loss": 1.0613, + "step": 1021 + }, + { + "epoch": 0.12288823423315097, + "grad_norm": 2.757077040687941, + "learning_rate": 3.910268818937608e-06, + "loss": 1.044, + "step": 1022 + }, + { + "epoch": 0.12300847712379005, + "grad_norm": 2.79092908956012, + "learning_rate": 3.9100379639253196e-06, + "loss": 1.1125, + "step": 1023 + }, + { + "epoch": 0.12312872001442915, + "grad_norm": 5.506633399665846, + "learning_rate": 3.909806819161791e-06, + "loss": 1.1082, + "step": 1024 + }, + { + "epoch": 0.12324896290506823, + "grad_norm": 2.285529420818939, + "learning_rate": 3.909575384682086e-06, + "loss": 1.1018, + "step": 1025 + }, + { + "epoch": 0.12336920579570733, + "grad_norm": 1.8646749406471208, + "learning_rate": 3.9093436605213144e-06, + "loss": 0.935, + "step": 1026 + }, + { + "epoch": 0.12348944868634643, + "grad_norm": 1.9210721952106395, + "learning_rate": 3.909111646714627e-06, + "loss": 1.0339, + "step": 1027 + }, + { + "epoch": 0.12360969157698551, + "grad_norm": 2.367474931948755, + "learning_rate": 3.9088793432972206e-06, + "loss": 0.9606, + "step": 1028 + }, + { + "epoch": 0.1237299344676246, + "grad_norm": 2.8133838536335105, + "learning_rate": 3.908646750304336e-06, + "loss": 1.0639, + "step": 1029 + }, + { + "epoch": 0.12385017735826369, + "grad_norm": 2.0081510147133574, + "learning_rate": 3.908413867771257e-06, + "loss": 1.1113, + "step": 1030 + }, + { + "epoch": 0.12397042024890279, + "grad_norm": 1.7029747152954646, + "learning_rate": 3.908180695733311e-06, + "loss": 1.0457, + "step": 1031 + }, + { + "epoch": 0.12409066313954187, + "grad_norm": 2.0145244107181712, + "learning_rate": 3.907947234225871e-06, + "loss": 1.0688, + "step": 1032 + }, + { + "epoch": 0.12421090603018096, + "grad_norm": 1.9360317696809914, + "learning_rate": 3.907713483284352e-06, + "loss": 1.1141, + "step": 1033 + }, + { + "epoch": 0.12433114892082006, + "grad_norm": 2.4499660489590966, + "learning_rate": 3.907479442944216e-06, + "loss": 1.2166, + "step": 1034 + }, + { + "epoch": 0.12445139181145914, + "grad_norm": 2.7678051992016237, + "learning_rate": 3.907245113240963e-06, + "loss": 1.1652, + "step": 1035 + }, + { + "epoch": 0.12457163470209824, + "grad_norm": 1.8645188162384358, + "learning_rate": 3.907010494210144e-06, + "loss": 0.9779, + "step": 1036 + }, + { + "epoch": 0.12469187759273732, + "grad_norm": 2.2983348125907597, + "learning_rate": 3.9067755858873495e-06, + "loss": 1.1582, + "step": 1037 + }, + { + "epoch": 0.12481212048337642, + "grad_norm": 0.8722135932324093, + "learning_rate": 3.906540388308214e-06, + "loss": 0.866, + "step": 1038 + }, + { + "epoch": 0.12493236337401552, + "grad_norm": 1.9237613259738704, + "learning_rate": 3.906304901508417e-06, + "loss": 1.0543, + "step": 1039 + }, + { + "epoch": 0.12505260626465461, + "grad_norm": 2.4747745064403444, + "learning_rate": 3.9060691255236835e-06, + "loss": 0.9993, + "step": 1040 + }, + { + "epoch": 0.1251728491552937, + "grad_norm": 2.1580697825394375, + "learning_rate": 3.905833060389778e-06, + "loss": 1.0438, + "step": 1041 + }, + { + "epoch": 0.12529309204593278, + "grad_norm": 9.46865935329445, + "learning_rate": 3.905596706142513e-06, + "loss": 1.022, + "step": 1042 + }, + { + "epoch": 0.12541333493657186, + "grad_norm": 4.789846433913475, + "learning_rate": 3.9053600628177435e-06, + "loss": 1.0984, + "step": 1043 + }, + { + "epoch": 0.12553357782721097, + "grad_norm": 2.204141260684533, + "learning_rate": 3.905123130451367e-06, + "loss": 1.0861, + "step": 1044 + }, + { + "epoch": 0.12565382071785006, + "grad_norm": 1.954767070952833, + "learning_rate": 3.904885909079326e-06, + "loss": 1.0309, + "step": 1045 + }, + { + "epoch": 0.12577406360848914, + "grad_norm": 3.027504406019386, + "learning_rate": 3.904648398737607e-06, + "loss": 1.0119, + "step": 1046 + }, + { + "epoch": 0.12589430649912825, + "grad_norm": 2.6212125430070863, + "learning_rate": 3.9044105994622406e-06, + "loss": 1.019, + "step": 1047 + }, + { + "epoch": 0.12601454938976733, + "grad_norm": 2.085805481579844, + "learning_rate": 3.9041725112893005e-06, + "loss": 1.0563, + "step": 1048 + }, + { + "epoch": 0.12613479228040642, + "grad_norm": 1.8899328052342976, + "learning_rate": 3.903934134254904e-06, + "loss": 0.9932, + "step": 1049 + }, + { + "epoch": 0.1262550351710455, + "grad_norm": 2.1188289571357966, + "learning_rate": 3.903695468395213e-06, + "loss": 1.0851, + "step": 1050 + }, + { + "epoch": 0.1263752780616846, + "grad_norm": 2.2783306724943944, + "learning_rate": 3.903456513746434e-06, + "loss": 0.7957, + "step": 1051 + }, + { + "epoch": 0.1264955209523237, + "grad_norm": 1.8497261566386585, + "learning_rate": 3.903217270344815e-06, + "loss": 1.1115, + "step": 1052 + }, + { + "epoch": 0.12661576384296278, + "grad_norm": 3.0470232022750725, + "learning_rate": 3.902977738226648e-06, + "loss": 1.0621, + "step": 1053 + }, + { + "epoch": 0.12673600673360189, + "grad_norm": 2.142906476769897, + "learning_rate": 3.902737917428273e-06, + "loss": 1.1463, + "step": 1054 + }, + { + "epoch": 0.12685624962424097, + "grad_norm": 1.9318293372255437, + "learning_rate": 3.902497807986068e-06, + "loss": 1.0749, + "step": 1055 + }, + { + "epoch": 0.12697649251488005, + "grad_norm": 1.7242909674305404, + "learning_rate": 3.902257409936458e-06, + "loss": 1.0807, + "step": 1056 + }, + { + "epoch": 0.12709673540551916, + "grad_norm": 2.139269434391363, + "learning_rate": 3.902016723315912e-06, + "loss": 1.0766, + "step": 1057 + }, + { + "epoch": 0.12721697829615825, + "grad_norm": 3.2921477501437955, + "learning_rate": 3.901775748160941e-06, + "loss": 0.9356, + "step": 1058 + }, + { + "epoch": 0.12733722118679733, + "grad_norm": 0.8248316578629962, + "learning_rate": 3.901534484508101e-06, + "loss": 0.8475, + "step": 1059 + }, + { + "epoch": 0.1274574640774364, + "grad_norm": 2.755508281531268, + "learning_rate": 3.901292932393991e-06, + "loss": 0.9842, + "step": 1060 + }, + { + "epoch": 0.12757770696807552, + "grad_norm": 2.5648605824852684, + "learning_rate": 3.9010510918552555e-06, + "loss": 1.0966, + "step": 1061 + }, + { + "epoch": 0.1276979498587146, + "grad_norm": 2.4364657041632, + "learning_rate": 3.900808962928581e-06, + "loss": 0.9881, + "step": 1062 + }, + { + "epoch": 0.1278181927493537, + "grad_norm": 2.609392542979951, + "learning_rate": 3.900566545650698e-06, + "loss": 1.1312, + "step": 1063 + }, + { + "epoch": 0.1279384356399928, + "grad_norm": 2.3597921040663863, + "learning_rate": 3.900323840058381e-06, + "loss": 1.0581, + "step": 1064 + }, + { + "epoch": 0.12805867853063188, + "grad_norm": 1.8866486151318056, + "learning_rate": 3.900080846188449e-06, + "loss": 1.0569, + "step": 1065 + }, + { + "epoch": 0.12817892142127096, + "grad_norm": 1.862031017372915, + "learning_rate": 3.8998375640777625e-06, + "loss": 1.0525, + "step": 1066 + }, + { + "epoch": 0.12829916431191005, + "grad_norm": 0.7095419632624346, + "learning_rate": 3.899593993763229e-06, + "loss": 0.7643, + "step": 1067 + }, + { + "epoch": 0.12841940720254916, + "grad_norm": 3.3253927701559802, + "learning_rate": 3.899350135281796e-06, + "loss": 1.0528, + "step": 1068 + }, + { + "epoch": 0.12853965009318824, + "grad_norm": 2.005368299832939, + "learning_rate": 3.8991059886704585e-06, + "loss": 1.0377, + "step": 1069 + }, + { + "epoch": 0.12865989298382732, + "grad_norm": 2.0250588773480325, + "learning_rate": 3.898861553966252e-06, + "loss": 1.0681, + "step": 1070 + }, + { + "epoch": 0.12878013587446643, + "grad_norm": 2.1817390877302287, + "learning_rate": 3.898616831206257e-06, + "loss": 1.1187, + "step": 1071 + }, + { + "epoch": 0.12890037876510552, + "grad_norm": 2.38566083921636, + "learning_rate": 3.8983718204276e-06, + "loss": 1.01, + "step": 1072 + }, + { + "epoch": 0.1290206216557446, + "grad_norm": 2.188989731466119, + "learning_rate": 3.898126521667446e-06, + "loss": 1.0694, + "step": 1073 + }, + { + "epoch": 0.12914086454638368, + "grad_norm": 1.8564092240294885, + "learning_rate": 3.897880934963007e-06, + "loss": 1.0742, + "step": 1074 + }, + { + "epoch": 0.1292611074370228, + "grad_norm": 2.278768783885315, + "learning_rate": 3.89763506035154e-06, + "loss": 1.0226, + "step": 1075 + }, + { + "epoch": 0.12938135032766188, + "grad_norm": 1.9144822886482846, + "learning_rate": 3.897388897870343e-06, + "loss": 1.0519, + "step": 1076 + }, + { + "epoch": 0.12950159321830096, + "grad_norm": 2.091521144328709, + "learning_rate": 3.89714244755676e-06, + "loss": 0.9863, + "step": 1077 + }, + { + "epoch": 0.12962183610894007, + "grad_norm": 3.173999845114153, + "learning_rate": 3.896895709448175e-06, + "loss": 1.0997, + "step": 1078 + }, + { + "epoch": 0.12974207899957915, + "grad_norm": 2.924116284804369, + "learning_rate": 3.896648683582019e-06, + "loss": 1.0128, + "step": 1079 + }, + { + "epoch": 0.12986232189021824, + "grad_norm": 2.251348727029698, + "learning_rate": 3.896401369995766e-06, + "loss": 1.0449, + "step": 1080 + }, + { + "epoch": 0.12998256478085732, + "grad_norm": 1.9365434438775249, + "learning_rate": 3.896153768726932e-06, + "loss": 1.034, + "step": 1081 + }, + { + "epoch": 0.13010280767149643, + "grad_norm": 2.7157294495823505, + "learning_rate": 3.8959058798130806e-06, + "loss": 1.1177, + "step": 1082 + }, + { + "epoch": 0.1302230505621355, + "grad_norm": 4.637379137299594, + "learning_rate": 3.895657703291814e-06, + "loss": 0.9896, + "step": 1083 + }, + { + "epoch": 0.1303432934527746, + "grad_norm": 3.2072498093727715, + "learning_rate": 3.895409239200781e-06, + "loss": 1.0361, + "step": 1084 + }, + { + "epoch": 0.1304635363434137, + "grad_norm": 2.5115433271264114, + "learning_rate": 3.895160487577673e-06, + "loss": 1.1578, + "step": 1085 + }, + { + "epoch": 0.1305837792340528, + "grad_norm": 0.7917857624284754, + "learning_rate": 3.894911448460226e-06, + "loss": 0.8471, + "step": 1086 + }, + { + "epoch": 0.13070402212469187, + "grad_norm": 2.085960816670947, + "learning_rate": 3.8946621218862195e-06, + "loss": 0.967, + "step": 1087 + }, + { + "epoch": 0.13082426501533098, + "grad_norm": 1.9228881398485875, + "learning_rate": 3.894412507893475e-06, + "loss": 1.129, + "step": 1088 + }, + { + "epoch": 0.13094450790597006, + "grad_norm": 2.5139261515778775, + "learning_rate": 3.894162606519859e-06, + "loss": 0.957, + "step": 1089 + }, + { + "epoch": 0.13106475079660915, + "grad_norm": 2.2617609155180047, + "learning_rate": 3.893912417803282e-06, + "loss": 1.0093, + "step": 1090 + }, + { + "epoch": 0.13118499368724823, + "grad_norm": 1.9686904171601396, + "learning_rate": 3.8936619417816975e-06, + "loss": 1.0075, + "step": 1091 + }, + { + "epoch": 0.13130523657788734, + "grad_norm": 2.725070927948007, + "learning_rate": 3.8934111784931015e-06, + "loss": 0.9522, + "step": 1092 + }, + { + "epoch": 0.13142547946852642, + "grad_norm": 1.065515726253572, + "learning_rate": 3.893160127975535e-06, + "loss": 0.8283, + "step": 1093 + }, + { + "epoch": 0.1315457223591655, + "grad_norm": 3.0862276737584446, + "learning_rate": 3.8929087902670826e-06, + "loss": 1.053, + "step": 1094 + }, + { + "epoch": 0.13166596524980462, + "grad_norm": 0.9251646732977044, + "learning_rate": 3.8926571654058715e-06, + "loss": 0.845, + "step": 1095 + }, + { + "epoch": 0.1317862081404437, + "grad_norm": 2.684243113927188, + "learning_rate": 3.892405253430074e-06, + "loss": 1.0124, + "step": 1096 + }, + { + "epoch": 0.13190645103108278, + "grad_norm": 2.1601917891657223, + "learning_rate": 3.892153054377904e-06, + "loss": 1.0605, + "step": 1097 + }, + { + "epoch": 0.13202669392172187, + "grad_norm": 0.9375137328095748, + "learning_rate": 3.891900568287619e-06, + "loss": 0.8325, + "step": 1098 + }, + { + "epoch": 0.13214693681236098, + "grad_norm": 2.4032015625820335, + "learning_rate": 3.891647795197523e-06, + "loss": 0.9585, + "step": 1099 + }, + { + "epoch": 0.13226717970300006, + "grad_norm": 2.343721109847985, + "learning_rate": 3.8913947351459605e-06, + "loss": 0.9267, + "step": 1100 + }, + { + "epoch": 0.13238742259363914, + "grad_norm": 1.9409091858095304, + "learning_rate": 3.89114138817132e-06, + "loss": 0.9117, + "step": 1101 + }, + { + "epoch": 0.13250766548427825, + "grad_norm": 1.8639145260807073, + "learning_rate": 3.890887754312035e-06, + "loss": 1.0809, + "step": 1102 + }, + { + "epoch": 0.13262790837491734, + "grad_norm": 2.5073732365735566, + "learning_rate": 3.890633833606581e-06, + "loss": 1.1153, + "step": 1103 + }, + { + "epoch": 0.13274815126555642, + "grad_norm": 1.8817906436111447, + "learning_rate": 3.890379626093477e-06, + "loss": 0.9331, + "step": 1104 + }, + { + "epoch": 0.1328683941561955, + "grad_norm": 2.4703354392292582, + "learning_rate": 3.890125131811287e-06, + "loss": 1.1595, + "step": 1105 + }, + { + "epoch": 0.1329886370468346, + "grad_norm": 30.006466995023548, + "learning_rate": 3.889870350798618e-06, + "loss": 0.9919, + "step": 1106 + }, + { + "epoch": 0.1331088799374737, + "grad_norm": 1.9118891619394425, + "learning_rate": 3.889615283094119e-06, + "loss": 1.0268, + "step": 1107 + }, + { + "epoch": 0.13322912282811278, + "grad_norm": 2.6492878946760303, + "learning_rate": 3.889359928736485e-06, + "loss": 1.0843, + "step": 1108 + }, + { + "epoch": 0.1333493657187519, + "grad_norm": 2.4184909928307703, + "learning_rate": 3.889104287764451e-06, + "loss": 1.151, + "step": 1109 + }, + { + "epoch": 0.13346960860939097, + "grad_norm": 2.1403467596171906, + "learning_rate": 3.888848360216798e-06, + "loss": 1.1459, + "step": 1110 + }, + { + "epoch": 0.13358985150003005, + "grad_norm": 0.8445770307725095, + "learning_rate": 3.888592146132351e-06, + "loss": 0.804, + "step": 1111 + }, + { + "epoch": 0.13371009439066917, + "grad_norm": 1.9930968359712464, + "learning_rate": 3.888335645549978e-06, + "loss": 1.0229, + "step": 1112 + }, + { + "epoch": 0.13383033728130825, + "grad_norm": 2.9234101041879272, + "learning_rate": 3.888078858508588e-06, + "loss": 1.0517, + "step": 1113 + }, + { + "epoch": 0.13395058017194733, + "grad_norm": 1.9383769358057457, + "learning_rate": 3.8878217850471365e-06, + "loss": 1.081, + "step": 1114 + }, + { + "epoch": 0.13407082306258641, + "grad_norm": 1.9337255567844114, + "learning_rate": 3.887564425204621e-06, + "loss": 0.9773, + "step": 1115 + }, + { + "epoch": 0.13419106595322552, + "grad_norm": 0.8434622591843125, + "learning_rate": 3.887306779020083e-06, + "loss": 0.7846, + "step": 1116 + }, + { + "epoch": 0.1343113088438646, + "grad_norm": 2.5918235518714705, + "learning_rate": 3.887048846532608e-06, + "loss": 0.9444, + "step": 1117 + }, + { + "epoch": 0.1344315517345037, + "grad_norm": 0.7606142693094138, + "learning_rate": 3.8867906277813224e-06, + "loss": 0.8197, + "step": 1118 + }, + { + "epoch": 0.1345517946251428, + "grad_norm": 2.3867744901301085, + "learning_rate": 3.886532122805399e-06, + "loss": 0.9768, + "step": 1119 + }, + { + "epoch": 0.13467203751578188, + "grad_norm": 3.470056923391129, + "learning_rate": 3.886273331644053e-06, + "loss": 1.1336, + "step": 1120 + }, + { + "epoch": 0.13479228040642097, + "grad_norm": 4.048342641163021, + "learning_rate": 3.886014254336542e-06, + "loss": 1.061, + "step": 1121 + }, + { + "epoch": 0.13491252329706005, + "grad_norm": 2.0406401512436423, + "learning_rate": 3.885754890922168e-06, + "loss": 1.1649, + "step": 1122 + }, + { + "epoch": 0.13503276618769916, + "grad_norm": 2.162259890064984, + "learning_rate": 3.885495241440277e-06, + "loss": 1.0226, + "step": 1123 + }, + { + "epoch": 0.13515300907833824, + "grad_norm": 1.9167194220981056, + "learning_rate": 3.885235305930257e-06, + "loss": 0.9807, + "step": 1124 + }, + { + "epoch": 0.13527325196897733, + "grad_norm": 1.938728004886648, + "learning_rate": 3.884975084431539e-06, + "loss": 1.092, + "step": 1125 + }, + { + "epoch": 0.13539349485961644, + "grad_norm": 4.0959822996621345, + "learning_rate": 3.8847145769836e-06, + "loss": 1.1525, + "step": 1126 + }, + { + "epoch": 0.13551373775025552, + "grad_norm": 3.1882827396104245, + "learning_rate": 3.884453783625959e-06, + "loss": 0.9035, + "step": 1127 + }, + { + "epoch": 0.1356339806408946, + "grad_norm": 2.383692444465058, + "learning_rate": 3.884192704398176e-06, + "loss": 1.085, + "step": 1128 + }, + { + "epoch": 0.13575422353153369, + "grad_norm": 1.8057623630667148, + "learning_rate": 3.883931339339858e-06, + "loss": 0.9817, + "step": 1129 + }, + { + "epoch": 0.1358744664221728, + "grad_norm": 2.0718523165236404, + "learning_rate": 3.883669688490654e-06, + "loss": 1.029, + "step": 1130 + }, + { + "epoch": 0.13599470931281188, + "grad_norm": 3.45188142642242, + "learning_rate": 3.883407751890256e-06, + "loss": 1.0934, + "step": 1131 + }, + { + "epoch": 0.13611495220345096, + "grad_norm": 2.3221844144823, + "learning_rate": 3.8831455295783994e-06, + "loss": 1.0991, + "step": 1132 + }, + { + "epoch": 0.13623519509409007, + "grad_norm": 1.964035201508902, + "learning_rate": 3.882883021594864e-06, + "loss": 0.9761, + "step": 1133 + }, + { + "epoch": 0.13635543798472916, + "grad_norm": 2.21997648914019, + "learning_rate": 3.8826202279794705e-06, + "loss": 1.1102, + "step": 1134 + }, + { + "epoch": 0.13647568087536824, + "grad_norm": 3.0850229187644227, + "learning_rate": 3.882357148772085e-06, + "loss": 0.9436, + "step": 1135 + }, + { + "epoch": 0.13659592376600732, + "grad_norm": 3.7432280111286755, + "learning_rate": 3.882093784012617e-06, + "loss": 1.0866, + "step": 1136 + }, + { + "epoch": 0.13671616665664643, + "grad_norm": 1.9372283991289838, + "learning_rate": 3.881830133741019e-06, + "loss": 1.0829, + "step": 1137 + }, + { + "epoch": 0.13683640954728551, + "grad_norm": 3.204707057378006, + "learning_rate": 3.881566197997285e-06, + "loss": 1.0058, + "step": 1138 + }, + { + "epoch": 0.1369566524379246, + "grad_norm": 1.6245357143600891, + "learning_rate": 3.881301976821456e-06, + "loss": 0.9895, + "step": 1139 + }, + { + "epoch": 0.1370768953285637, + "grad_norm": 1.8332005004733969, + "learning_rate": 3.881037470253612e-06, + "loss": 1.1434, + "step": 1140 + }, + { + "epoch": 0.1371971382192028, + "grad_norm": 3.685797346748574, + "learning_rate": 3.88077267833388e-06, + "loss": 1.0362, + "step": 1141 + }, + { + "epoch": 0.13731738110984187, + "grad_norm": 2.99433411572086, + "learning_rate": 3.880507601102427e-06, + "loss": 1.0815, + "step": 1142 + }, + { + "epoch": 0.13743762400048098, + "grad_norm": 1.9692609214852759, + "learning_rate": 3.880242238599467e-06, + "loss": 1.0615, + "step": 1143 + }, + { + "epoch": 0.13755786689112007, + "grad_norm": 1.7502435787166504, + "learning_rate": 3.879976590865254e-06, + "loss": 1.0708, + "step": 1144 + }, + { + "epoch": 0.13767810978175915, + "grad_norm": 2.037771466664995, + "learning_rate": 3.879710657940087e-06, + "loss": 1.1117, + "step": 1145 + }, + { + "epoch": 0.13779835267239823, + "grad_norm": 2.484785130074462, + "learning_rate": 3.879444439864308e-06, + "loss": 0.9431, + "step": 1146 + }, + { + "epoch": 0.13791859556303734, + "grad_norm": 4.351141917862469, + "learning_rate": 3.879177936678301e-06, + "loss": 1.1023, + "step": 1147 + }, + { + "epoch": 0.13803883845367643, + "grad_norm": 2.365549962656459, + "learning_rate": 3.878911148422496e-06, + "loss": 1.0093, + "step": 1148 + }, + { + "epoch": 0.1381590813443155, + "grad_norm": 2.4434191896967596, + "learning_rate": 3.878644075137364e-06, + "loss": 0.9493, + "step": 1149 + }, + { + "epoch": 0.13827932423495462, + "grad_norm": 3.458674050211715, + "learning_rate": 3.878376716863418e-06, + "loss": 1.0312, + "step": 1150 + }, + { + "epoch": 0.1383995671255937, + "grad_norm": 2.236685245434859, + "learning_rate": 3.878109073641219e-06, + "loss": 0.9533, + "step": 1151 + }, + { + "epoch": 0.13851981001623279, + "grad_norm": 1.6833335870169772, + "learning_rate": 3.877841145511366e-06, + "loss": 1.053, + "step": 1152 + }, + { + "epoch": 0.13864005290687187, + "grad_norm": 1.7859511327348365, + "learning_rate": 3.8775729325145035e-06, + "loss": 1.0687, + "step": 1153 + }, + { + "epoch": 0.13876029579751098, + "grad_norm": 0.7925337928041373, + "learning_rate": 3.877304434691321e-06, + "loss": 0.8848, + "step": 1154 + }, + { + "epoch": 0.13888053868815006, + "grad_norm": 2.099802975267591, + "learning_rate": 3.877035652082548e-06, + "loss": 1.0369, + "step": 1155 + }, + { + "epoch": 0.13900078157878915, + "grad_norm": 1.805540535530151, + "learning_rate": 3.87676658472896e-06, + "loss": 1.0961, + "step": 1156 + }, + { + "epoch": 0.13912102446942826, + "grad_norm": 2.5490525187083817, + "learning_rate": 3.876497232671372e-06, + "loss": 1.0904, + "step": 1157 + }, + { + "epoch": 0.13924126736006734, + "grad_norm": 5.70734940933823, + "learning_rate": 3.876227595950647e-06, + "loss": 1.0739, + "step": 1158 + }, + { + "epoch": 0.13936151025070642, + "grad_norm": 3.4738421480270487, + "learning_rate": 3.875957674607686e-06, + "loss": 1.0283, + "step": 1159 + }, + { + "epoch": 0.1394817531413455, + "grad_norm": 2.2914692071172915, + "learning_rate": 3.8756874686834386e-06, + "loss": 1.1151, + "step": 1160 + }, + { + "epoch": 0.13960199603198462, + "grad_norm": 1.6371538043791474, + "learning_rate": 3.875416978218893e-06, + "loss": 1.0442, + "step": 1161 + }, + { + "epoch": 0.1397222389226237, + "grad_norm": 4.892381054227148, + "learning_rate": 3.8751462032550835e-06, + "loss": 1.0669, + "step": 1162 + }, + { + "epoch": 0.13984248181326278, + "grad_norm": 2.297994431683971, + "learning_rate": 3.874875143833085e-06, + "loss": 1.0644, + "step": 1163 + }, + { + "epoch": 0.1399627247039019, + "grad_norm": 2.02103803194361, + "learning_rate": 3.874603799994019e-06, + "loss": 0.928, + "step": 1164 + }, + { + "epoch": 0.14008296759454097, + "grad_norm": 2.2130805638572677, + "learning_rate": 3.874332171779046e-06, + "loss": 1.1081, + "step": 1165 + }, + { + "epoch": 0.14020321048518006, + "grad_norm": 1.8113506224425728, + "learning_rate": 3.874060259229373e-06, + "loss": 0.9951, + "step": 1166 + }, + { + "epoch": 0.14032345337581917, + "grad_norm": 2.214556640100482, + "learning_rate": 3.873788062386249e-06, + "loss": 1.1808, + "step": 1167 + }, + { + "epoch": 0.14044369626645825, + "grad_norm": 2.072315211875796, + "learning_rate": 3.873515581290965e-06, + "loss": 1.0596, + "step": 1168 + }, + { + "epoch": 0.14056393915709733, + "grad_norm": 2.2740097112322464, + "learning_rate": 3.8732428159848575e-06, + "loss": 0.9923, + "step": 1169 + }, + { + "epoch": 0.14068418204773642, + "grad_norm": 2.1652182236918467, + "learning_rate": 3.872969766509304e-06, + "loss": 1.0221, + "step": 1170 + }, + { + "epoch": 0.14080442493837553, + "grad_norm": 0.7657598259993155, + "learning_rate": 3.872696432905726e-06, + "loss": 0.794, + "step": 1171 + }, + { + "epoch": 0.1409246678290146, + "grad_norm": 2.2080896591005885, + "learning_rate": 3.872422815215589e-06, + "loss": 0.9532, + "step": 1172 + }, + { + "epoch": 0.1410449107196537, + "grad_norm": 2.009758981706707, + "learning_rate": 3.8721489134803994e-06, + "loss": 0.9847, + "step": 1173 + }, + { + "epoch": 0.1411651536102928, + "grad_norm": 2.2419768443821737, + "learning_rate": 3.871874727741707e-06, + "loss": 0.9658, + "step": 1174 + }, + { + "epoch": 0.1412853965009319, + "grad_norm": 1.834293424751822, + "learning_rate": 3.871600258041108e-06, + "loss": 1.2053, + "step": 1175 + }, + { + "epoch": 0.14140563939157097, + "grad_norm": 2.316704819271331, + "learning_rate": 3.871325504420238e-06, + "loss": 1.0964, + "step": 1176 + }, + { + "epoch": 0.14152588228221005, + "grad_norm": 2.361038531016868, + "learning_rate": 3.871050466920776e-06, + "loss": 1.0563, + "step": 1177 + }, + { + "epoch": 0.14164612517284916, + "grad_norm": 1.8484930034955955, + "learning_rate": 3.870775145584447e-06, + "loss": 1.0341, + "step": 1178 + }, + { + "epoch": 0.14176636806348825, + "grad_norm": 2.717072769734951, + "learning_rate": 3.8704995404530145e-06, + "loss": 0.8848, + "step": 1179 + }, + { + "epoch": 0.14188661095412733, + "grad_norm": 2.317205024390599, + "learning_rate": 3.87022365156829e-06, + "loss": 1.0926, + "step": 1180 + }, + { + "epoch": 0.14200685384476644, + "grad_norm": 2.0561725775801807, + "learning_rate": 3.869947478972123e-06, + "loss": 1.0509, + "step": 1181 + }, + { + "epoch": 0.14212709673540552, + "grad_norm": 2.3607286018743796, + "learning_rate": 3.869671022706412e-06, + "loss": 1.0587, + "step": 1182 + }, + { + "epoch": 0.1422473396260446, + "grad_norm": 2.3178928487531043, + "learning_rate": 3.869394282813092e-06, + "loss": 0.8922, + "step": 1183 + }, + { + "epoch": 0.1423675825166837, + "grad_norm": 3.2817165951266998, + "learning_rate": 3.869117259334147e-06, + "loss": 1.131, + "step": 1184 + }, + { + "epoch": 0.1424878254073228, + "grad_norm": 1.6887218855318344, + "learning_rate": 3.868839952311599e-06, + "loss": 1.0584, + "step": 1185 + }, + { + "epoch": 0.14260806829796188, + "grad_norm": 3.416789556634797, + "learning_rate": 3.868562361787516e-06, + "loss": 1.0445, + "step": 1186 + }, + { + "epoch": 0.14272831118860096, + "grad_norm": 2.1418213111865176, + "learning_rate": 3.868284487804009e-06, + "loss": 0.9352, + "step": 1187 + }, + { + "epoch": 0.14284855407924008, + "grad_norm": 1.8153503630464267, + "learning_rate": 3.86800633040323e-06, + "loss": 1.0239, + "step": 1188 + }, + { + "epoch": 0.14296879696987916, + "grad_norm": 2.1690928251418913, + "learning_rate": 3.867727889627376e-06, + "loss": 1.0243, + "step": 1189 + }, + { + "epoch": 0.14308903986051824, + "grad_norm": 2.461048523351287, + "learning_rate": 3.867449165518687e-06, + "loss": 1.0195, + "step": 1190 + }, + { + "epoch": 0.14320928275115732, + "grad_norm": 2.3887359535686614, + "learning_rate": 3.867170158119444e-06, + "loss": 0.9515, + "step": 1191 + }, + { + "epoch": 0.14332952564179643, + "grad_norm": 1.9944135608011615, + "learning_rate": 3.866890867471972e-06, + "loss": 0.9891, + "step": 1192 + }, + { + "epoch": 0.14344976853243552, + "grad_norm": 3.1420547495247804, + "learning_rate": 3.86661129361864e-06, + "loss": 1.1365, + "step": 1193 + }, + { + "epoch": 0.1435700114230746, + "grad_norm": 2.1637376892691655, + "learning_rate": 3.866331436601859e-06, + "loss": 1.0981, + "step": 1194 + }, + { + "epoch": 0.1436902543137137, + "grad_norm": 3.361372810523356, + "learning_rate": 3.866051296464083e-06, + "loss": 0.9759, + "step": 1195 + }, + { + "epoch": 0.1438104972043528, + "grad_norm": 2.6567116728846956, + "learning_rate": 3.86577087324781e-06, + "loss": 1.0931, + "step": 1196 + }, + { + "epoch": 0.14393074009499188, + "grad_norm": 2.4640638081723103, + "learning_rate": 3.865490166995578e-06, + "loss": 1.0191, + "step": 1197 + }, + { + "epoch": 0.144050982985631, + "grad_norm": 2.922032642701763, + "learning_rate": 3.86520917774997e-06, + "loss": 1.0784, + "step": 1198 + }, + { + "epoch": 0.14417122587627007, + "grad_norm": 2.8546543864817973, + "learning_rate": 3.864927905553614e-06, + "loss": 0.9898, + "step": 1199 + }, + { + "epoch": 0.14429146876690915, + "grad_norm": 3.3666236918762746, + "learning_rate": 3.8646463504491765e-06, + "loss": 1.1299, + "step": 1200 + }, + { + "epoch": 0.14441171165754824, + "grad_norm": 1.7625318456708192, + "learning_rate": 3.8643645124793705e-06, + "loss": 1.0738, + "step": 1201 + }, + { + "epoch": 0.14453195454818735, + "grad_norm": 1.9219393215425737, + "learning_rate": 3.8640823916869515e-06, + "loss": 0.9846, + "step": 1202 + }, + { + "epoch": 0.14465219743882643, + "grad_norm": 1.6798072949650424, + "learning_rate": 3.863799988114714e-06, + "loss": 1.024, + "step": 1203 + }, + { + "epoch": 0.1447724403294655, + "grad_norm": 2.641736580999535, + "learning_rate": 3.863517301805502e-06, + "loss": 0.9427, + "step": 1204 + }, + { + "epoch": 0.14489268322010462, + "grad_norm": 2.5236387379245118, + "learning_rate": 3.863234332802196e-06, + "loss": 1.2077, + "step": 1205 + }, + { + "epoch": 0.1450129261107437, + "grad_norm": 2.91978322596197, + "learning_rate": 3.862951081147723e-06, + "loss": 0.9759, + "step": 1206 + }, + { + "epoch": 0.1451331690013828, + "grad_norm": 2.2995874905731544, + "learning_rate": 3.862667546885053e-06, + "loss": 1.0224, + "step": 1207 + }, + { + "epoch": 0.14525341189202187, + "grad_norm": 2.2741009245647392, + "learning_rate": 3.8623837300571965e-06, + "loss": 0.9737, + "step": 1208 + }, + { + "epoch": 0.14537365478266098, + "grad_norm": 2.147958931073888, + "learning_rate": 3.8620996307072085e-06, + "loss": 1.0812, + "step": 1209 + }, + { + "epoch": 0.14549389767330007, + "grad_norm": 1.941159760391336, + "learning_rate": 3.861815248878188e-06, + "loss": 0.885, + "step": 1210 + }, + { + "epoch": 0.14561414056393915, + "grad_norm": 9.992246292096391, + "learning_rate": 3.861530584613274e-06, + "loss": 1.0396, + "step": 1211 + }, + { + "epoch": 0.14573438345457826, + "grad_norm": 2.28604739818679, + "learning_rate": 3.86124563795565e-06, + "loss": 1.0603, + "step": 1212 + }, + { + "epoch": 0.14585462634521734, + "grad_norm": 1.720610010959617, + "learning_rate": 3.860960408948543e-06, + "loss": 0.9405, + "step": 1213 + }, + { + "epoch": 0.14597486923585642, + "grad_norm": 2.770030677432735, + "learning_rate": 3.860674897635222e-06, + "loss": 1.1347, + "step": 1214 + }, + { + "epoch": 0.1460951121264955, + "grad_norm": 2.059915715317359, + "learning_rate": 3.860389104058998e-06, + "loss": 1.0727, + "step": 1215 + }, + { + "epoch": 0.14621535501713462, + "grad_norm": 1.9597097253915412, + "learning_rate": 3.860103028263227e-06, + "loss": 0.9667, + "step": 1216 + }, + { + "epoch": 0.1463355979077737, + "grad_norm": 2.4310924724156493, + "learning_rate": 3.859816670291304e-06, + "loss": 0.9413, + "step": 1217 + }, + { + "epoch": 0.14645584079841278, + "grad_norm": 2.747789708449384, + "learning_rate": 3.859530030186672e-06, + "loss": 1.1362, + "step": 1218 + }, + { + "epoch": 0.1465760836890519, + "grad_norm": 2.472238225087391, + "learning_rate": 3.859243107992813e-06, + "loss": 1.0625, + "step": 1219 + }, + { + "epoch": 0.14669632657969098, + "grad_norm": 4.402834170221353, + "learning_rate": 3.858955903753252e-06, + "loss": 1.0225, + "step": 1220 + }, + { + "epoch": 0.14681656947033006, + "grad_norm": 2.069754816913402, + "learning_rate": 3.858668417511559e-06, + "loss": 1.0729, + "step": 1221 + }, + { + "epoch": 0.14693681236096917, + "grad_norm": 2.6809453653348476, + "learning_rate": 3.8583806493113445e-06, + "loss": 1.0001, + "step": 1222 + }, + { + "epoch": 0.14705705525160825, + "grad_norm": 2.3115488106576367, + "learning_rate": 3.858092599196263e-06, + "loss": 1.0598, + "step": 1223 + }, + { + "epoch": 0.14717729814224734, + "grad_norm": 2.743288085567438, + "learning_rate": 3.857804267210012e-06, + "loss": 1.0641, + "step": 1224 + }, + { + "epoch": 0.14729754103288642, + "grad_norm": 2.693606336971269, + "learning_rate": 3.857515653396331e-06, + "loss": 1.1217, + "step": 1225 + }, + { + "epoch": 0.14741778392352553, + "grad_norm": 4.120070777951761, + "learning_rate": 3.857226757799002e-06, + "loss": 1.1078, + "step": 1226 + }, + { + "epoch": 0.1475380268141646, + "grad_norm": 2.81614245741477, + "learning_rate": 3.85693758046185e-06, + "loss": 0.9791, + "step": 1227 + }, + { + "epoch": 0.1476582697048037, + "grad_norm": 1.969571865439259, + "learning_rate": 3.8566481214287435e-06, + "loss": 1.0671, + "step": 1228 + }, + { + "epoch": 0.1477785125954428, + "grad_norm": 2.1119228093678686, + "learning_rate": 3.8563583807435935e-06, + "loss": 1.1481, + "step": 1229 + }, + { + "epoch": 0.1478987554860819, + "grad_norm": 4.240681024236132, + "learning_rate": 3.856068358450353e-06, + "loss": 1.0152, + "step": 1230 + }, + { + "epoch": 0.14801899837672097, + "grad_norm": 1.9247122190299968, + "learning_rate": 3.8557780545930186e-06, + "loss": 1.0976, + "step": 1231 + }, + { + "epoch": 0.14813924126736006, + "grad_norm": 1.7637402763071626, + "learning_rate": 3.855487469215628e-06, + "loss": 1.0322, + "step": 1232 + }, + { + "epoch": 0.14825948415799917, + "grad_norm": 2.3780346606385883, + "learning_rate": 3.855196602362264e-06, + "loss": 0.9594, + "step": 1233 + }, + { + "epoch": 0.14837972704863825, + "grad_norm": 2.9699123315684166, + "learning_rate": 3.854905454077051e-06, + "loss": 1.1795, + "step": 1234 + }, + { + "epoch": 0.14849996993927733, + "grad_norm": 2.213937725462887, + "learning_rate": 3.854614024404155e-06, + "loss": 1.1183, + "step": 1235 + }, + { + "epoch": 0.14862021282991644, + "grad_norm": 2.0343973278689638, + "learning_rate": 3.8543223133877865e-06, + "loss": 1.1322, + "step": 1236 + }, + { + "epoch": 0.14874045572055553, + "grad_norm": 2.2335143665938775, + "learning_rate": 3.854030321072198e-06, + "loss": 1.1204, + "step": 1237 + }, + { + "epoch": 0.1488606986111946, + "grad_norm": 2.462225489153265, + "learning_rate": 3.853738047501682e-06, + "loss": 0.9721, + "step": 1238 + }, + { + "epoch": 0.1489809415018337, + "grad_norm": 2.0232258689701745, + "learning_rate": 3.85344549272058e-06, + "loss": 1.0162, + "step": 1239 + }, + { + "epoch": 0.1491011843924728, + "grad_norm": 2.3200507738871377, + "learning_rate": 3.853152656773269e-06, + "loss": 1.0645, + "step": 1240 + }, + { + "epoch": 0.14922142728311188, + "grad_norm": 1.7090924909997551, + "learning_rate": 3.852859539704174e-06, + "loss": 1.0873, + "step": 1241 + }, + { + "epoch": 0.14934167017375097, + "grad_norm": 2.0214898003399417, + "learning_rate": 3.85256614155776e-06, + "loss": 1.0018, + "step": 1242 + }, + { + "epoch": 0.14946191306439008, + "grad_norm": 2.624483057619665, + "learning_rate": 3.852272462378535e-06, + "loss": 0.9808, + "step": 1243 + }, + { + "epoch": 0.14958215595502916, + "grad_norm": 3.2909646231211247, + "learning_rate": 3.85197850221105e-06, + "loss": 1.0182, + "step": 1244 + }, + { + "epoch": 0.14970239884566824, + "grad_norm": 1.8752118308888643, + "learning_rate": 3.851684261099899e-06, + "loss": 0.9924, + "step": 1245 + }, + { + "epoch": 0.14982264173630733, + "grad_norm": 2.2153093766922147, + "learning_rate": 3.851389739089718e-06, + "loss": 1.1052, + "step": 1246 + }, + { + "epoch": 0.14994288462694644, + "grad_norm": 6.407020485297627, + "learning_rate": 3.851094936225186e-06, + "loss": 1.0414, + "step": 1247 + }, + { + "epoch": 0.15006312751758552, + "grad_norm": 1.9281153896593077, + "learning_rate": 3.850799852551024e-06, + "loss": 1.0075, + "step": 1248 + }, + { + "epoch": 0.1501833704082246, + "grad_norm": 2.52364686268678, + "learning_rate": 3.850504488111995e-06, + "loss": 1.0929, + "step": 1249 + }, + { + "epoch": 0.15030361329886371, + "grad_norm": 2.000979064671752, + "learning_rate": 3.850208842952907e-06, + "loss": 1.0639, + "step": 1250 + }, + { + "epoch": 0.1504238561895028, + "grad_norm": 1.7379914803027252, + "learning_rate": 3.849912917118608e-06, + "loss": 1.0286, + "step": 1251 + }, + { + "epoch": 0.15054409908014188, + "grad_norm": 1.1507926199688652, + "learning_rate": 3.849616710653992e-06, + "loss": 0.8308, + "step": 1252 + }, + { + "epoch": 0.150664341970781, + "grad_norm": 24.30477037112198, + "learning_rate": 3.84932022360399e-06, + "loss": 0.9866, + "step": 1253 + }, + { + "epoch": 0.15078458486142007, + "grad_norm": 3.584243141416267, + "learning_rate": 3.849023456013581e-06, + "loss": 1.079, + "step": 1254 + }, + { + "epoch": 0.15090482775205916, + "grad_norm": 3.019215714128817, + "learning_rate": 3.848726407927784e-06, + "loss": 0.8623, + "step": 1255 + }, + { + "epoch": 0.15102507064269824, + "grad_norm": 3.0440858252277545, + "learning_rate": 3.84842907939166e-06, + "loss": 1.1017, + "step": 1256 + }, + { + "epoch": 0.15114531353333735, + "grad_norm": 2.588484049001661, + "learning_rate": 3.8481314704503146e-06, + "loss": 0.9447, + "step": 1257 + }, + { + "epoch": 0.15126555642397643, + "grad_norm": 2.3794601121394012, + "learning_rate": 3.847833581148895e-06, + "loss": 1.1166, + "step": 1258 + }, + { + "epoch": 0.15138579931461552, + "grad_norm": 6.65147321630822, + "learning_rate": 3.84753541153259e-06, + "loss": 1.053, + "step": 1259 + }, + { + "epoch": 0.15150604220525463, + "grad_norm": 1.7247060055300656, + "learning_rate": 3.847236961646633e-06, + "loss": 1.0717, + "step": 1260 + }, + { + "epoch": 0.1516262850958937, + "grad_norm": 2.40337874284543, + "learning_rate": 3.846938231536296e-06, + "loss": 1.0232, + "step": 1261 + }, + { + "epoch": 0.1517465279865328, + "grad_norm": 2.1935125282195904, + "learning_rate": 3.8466392212468995e-06, + "loss": 1.0488, + "step": 1262 + }, + { + "epoch": 0.15186677087717187, + "grad_norm": 0.8249721609100215, + "learning_rate": 3.8463399308238e-06, + "loss": 0.8574, + "step": 1263 + }, + { + "epoch": 0.15198701376781099, + "grad_norm": 1.9420257132700889, + "learning_rate": 3.846040360312402e-06, + "loss": 0.8771, + "step": 1264 + }, + { + "epoch": 0.15210725665845007, + "grad_norm": 2.393298082047803, + "learning_rate": 3.8457405097581485e-06, + "loss": 1.0535, + "step": 1265 + }, + { + "epoch": 0.15222749954908915, + "grad_norm": 1.8358284126500928, + "learning_rate": 3.8454403792065275e-06, + "loss": 1.0136, + "step": 1266 + }, + { + "epoch": 0.15234774243972826, + "grad_norm": 2.135447122000017, + "learning_rate": 3.845139968703068e-06, + "loss": 1.0895, + "step": 1267 + }, + { + "epoch": 0.15246798533036734, + "grad_norm": 2.0173901066873046, + "learning_rate": 3.844839278293342e-06, + "loss": 1.0712, + "step": 1268 + }, + { + "epoch": 0.15258822822100643, + "grad_norm": 3.3437092145767564, + "learning_rate": 3.8445383080229654e-06, + "loss": 1.006, + "step": 1269 + }, + { + "epoch": 0.1527084711116455, + "grad_norm": 2.5410990367806963, + "learning_rate": 3.844237057937593e-06, + "loss": 0.9721, + "step": 1270 + }, + { + "epoch": 0.15282871400228462, + "grad_norm": 2.6302947278831295, + "learning_rate": 3.843935528082926e-06, + "loss": 1.014, + "step": 1271 + }, + { + "epoch": 0.1529489568929237, + "grad_norm": 2.087244545844817, + "learning_rate": 3.843633718504704e-06, + "loss": 1.0878, + "step": 1272 + }, + { + "epoch": 0.1530691997835628, + "grad_norm": 3.031332899710084, + "learning_rate": 3.843331629248715e-06, + "loss": 1.1357, + "step": 1273 + }, + { + "epoch": 0.1531894426742019, + "grad_norm": 2.1915916323659936, + "learning_rate": 3.843029260360782e-06, + "loss": 1.0043, + "step": 1274 + }, + { + "epoch": 0.15330968556484098, + "grad_norm": 1.991782173361333, + "learning_rate": 3.8427266118867755e-06, + "loss": 1.0296, + "step": 1275 + }, + { + "epoch": 0.15342992845548006, + "grad_norm": 2.1857495525018886, + "learning_rate": 3.842423683872608e-06, + "loss": 1.0652, + "step": 1276 + }, + { + "epoch": 0.15355017134611917, + "grad_norm": 3.55751572327346, + "learning_rate": 3.842120476364232e-06, + "loss": 1.0126, + "step": 1277 + }, + { + "epoch": 0.15367041423675826, + "grad_norm": 2.3022540160090363, + "learning_rate": 3.841816989407644e-06, + "loss": 1.074, + "step": 1278 + }, + { + "epoch": 0.15379065712739734, + "grad_norm": 2.3717801201983533, + "learning_rate": 3.841513223048884e-06, + "loss": 1.0013, + "step": 1279 + }, + { + "epoch": 0.15391090001803642, + "grad_norm": 5.256190328905684, + "learning_rate": 3.841209177334031e-06, + "loss": 1.0224, + "step": 1280 + }, + { + "epoch": 0.15403114290867553, + "grad_norm": 2.11357422100607, + "learning_rate": 3.84090485230921e-06, + "loss": 0.9883, + "step": 1281 + }, + { + "epoch": 0.15415138579931462, + "grad_norm": 3.8947899521100147, + "learning_rate": 3.840600248020588e-06, + "loss": 1.007, + "step": 1282 + }, + { + "epoch": 0.1542716286899537, + "grad_norm": 2.39838898948295, + "learning_rate": 3.840295364514371e-06, + "loss": 1.0376, + "step": 1283 + }, + { + "epoch": 0.1543918715805928, + "grad_norm": 2.164041567263945, + "learning_rate": 3.83999020183681e-06, + "loss": 1.0222, + "step": 1284 + }, + { + "epoch": 0.1545121144712319, + "grad_norm": 3.3891169275017363, + "learning_rate": 3.839684760034199e-06, + "loss": 1.0206, + "step": 1285 + }, + { + "epoch": 0.15463235736187098, + "grad_norm": 2.5708369036072147, + "learning_rate": 3.8393790391528716e-06, + "loss": 0.8908, + "step": 1286 + }, + { + "epoch": 0.15475260025251006, + "grad_norm": 1.8595146319107798, + "learning_rate": 3.8390730392392075e-06, + "loss": 1.1252, + "step": 1287 + }, + { + "epoch": 0.15487284314314917, + "grad_norm": 2.5196550205791053, + "learning_rate": 3.838766760339626e-06, + "loss": 1.0341, + "step": 1288 + }, + { + "epoch": 0.15499308603378825, + "grad_norm": 2.590759575710989, + "learning_rate": 3.838460202500587e-06, + "loss": 1.0316, + "step": 1289 + }, + { + "epoch": 0.15511332892442733, + "grad_norm": 2.5447817198071907, + "learning_rate": 3.838153365768599e-06, + "loss": 0.9831, + "step": 1290 + }, + { + "epoch": 0.15523357181506645, + "grad_norm": 4.917914259217087, + "learning_rate": 3.837846250190206e-06, + "loss": 0.9882, + "step": 1291 + }, + { + "epoch": 0.15535381470570553, + "grad_norm": 2.5375680134495076, + "learning_rate": 3.837538855811998e-06, + "loss": 1.0089, + "step": 1292 + }, + { + "epoch": 0.1554740575963446, + "grad_norm": 2.4128821984382163, + "learning_rate": 3.837231182680606e-06, + "loss": 0.9498, + "step": 1293 + }, + { + "epoch": 0.1555943004869837, + "grad_norm": 2.781562959015057, + "learning_rate": 3.836923230842706e-06, + "loss": 1.0007, + "step": 1294 + }, + { + "epoch": 0.1557145433776228, + "grad_norm": 2.40375429007256, + "learning_rate": 3.836615000345011e-06, + "loss": 1.0414, + "step": 1295 + }, + { + "epoch": 0.1558347862682619, + "grad_norm": 2.2556932264301137, + "learning_rate": 3.836306491234282e-06, + "loss": 1.0203, + "step": 1296 + }, + { + "epoch": 0.15595502915890097, + "grad_norm": 3.783014059389042, + "learning_rate": 3.835997703557317e-06, + "loss": 0.9902, + "step": 1297 + }, + { + "epoch": 0.15607527204954008, + "grad_norm": 1.7546432068313544, + "learning_rate": 3.83568863736096e-06, + "loss": 1.0404, + "step": 1298 + }, + { + "epoch": 0.15619551494017916, + "grad_norm": 2.8581258411185178, + "learning_rate": 3.8353792926920975e-06, + "loss": 1.1357, + "step": 1299 + }, + { + "epoch": 0.15631575783081825, + "grad_norm": 2.5310012318379433, + "learning_rate": 3.835069669597655e-06, + "loss": 1.0562, + "step": 1300 + }, + { + "epoch": 0.15643600072145733, + "grad_norm": 1.99338893191944, + "learning_rate": 3.834759768124603e-06, + "loss": 1.0345, + "step": 1301 + }, + { + "epoch": 0.15655624361209644, + "grad_norm": 2.9888985588393155, + "learning_rate": 3.834449588319953e-06, + "loss": 1.0026, + "step": 1302 + }, + { + "epoch": 0.15667648650273552, + "grad_norm": 1.9365315016339597, + "learning_rate": 3.834139130230758e-06, + "loss": 1.0863, + "step": 1303 + }, + { + "epoch": 0.1567967293933746, + "grad_norm": 1.7321643659466954, + "learning_rate": 3.833828393904117e-06, + "loss": 1.0489, + "step": 1304 + }, + { + "epoch": 0.15691697228401372, + "grad_norm": 2.569587025395151, + "learning_rate": 3.833517379387165e-06, + "loss": 1.0126, + "step": 1305 + }, + { + "epoch": 0.1570372151746528, + "grad_norm": 2.1384901553077453, + "learning_rate": 3.833206086727085e-06, + "loss": 1.1276, + "step": 1306 + }, + { + "epoch": 0.15715745806529188, + "grad_norm": 2.2536206254518336, + "learning_rate": 3.8328945159710994e-06, + "loss": 0.946, + "step": 1307 + }, + { + "epoch": 0.157277700955931, + "grad_norm": 2.4484371416570014, + "learning_rate": 3.832582667166473e-06, + "loss": 1.1282, + "step": 1308 + }, + { + "epoch": 0.15739794384657008, + "grad_norm": 1.8453370069254065, + "learning_rate": 3.8322705403605125e-06, + "loss": 1.0521, + "step": 1309 + }, + { + "epoch": 0.15751818673720916, + "grad_norm": 2.3209961163489554, + "learning_rate": 3.831958135600568e-06, + "loss": 1.0531, + "step": 1310 + }, + { + "epoch": 0.15763842962784824, + "grad_norm": 2.359536171770277, + "learning_rate": 3.831645452934032e-06, + "loss": 1.0363, + "step": 1311 + }, + { + "epoch": 0.15775867251848735, + "grad_norm": 1.8105909224895844, + "learning_rate": 3.831332492408336e-06, + "loss": 1.0399, + "step": 1312 + }, + { + "epoch": 0.15787891540912644, + "grad_norm": 2.9380239465826956, + "learning_rate": 3.831019254070957e-06, + "loss": 0.9317, + "step": 1313 + }, + { + "epoch": 0.15799915829976552, + "grad_norm": 2.8044248155080576, + "learning_rate": 3.8307057379694135e-06, + "loss": 1.1886, + "step": 1314 + }, + { + "epoch": 0.15811940119040463, + "grad_norm": 2.3233507603229357, + "learning_rate": 3.830391944151264e-06, + "loss": 1.0606, + "step": 1315 + }, + { + "epoch": 0.1582396440810437, + "grad_norm": 2.400739528046443, + "learning_rate": 3.830077872664114e-06, + "loss": 0.9127, + "step": 1316 + }, + { + "epoch": 0.1583598869716828, + "grad_norm": 2.451778941571307, + "learning_rate": 3.829763523555604e-06, + "loss": 0.9684, + "step": 1317 + }, + { + "epoch": 0.15848012986232188, + "grad_norm": 4.896887915910097, + "learning_rate": 3.829448896873423e-06, + "loss": 1.0169, + "step": 1318 + }, + { + "epoch": 0.158600372752961, + "grad_norm": 2.0276720666838033, + "learning_rate": 3.829133992665299e-06, + "loss": 1.0315, + "step": 1319 + }, + { + "epoch": 0.15872061564360007, + "grad_norm": 2.6764751787522347, + "learning_rate": 3.828818810979002e-06, + "loss": 1.1279, + "step": 1320 + }, + { + "epoch": 0.15884085853423915, + "grad_norm": 1.7893697150042769, + "learning_rate": 3.8285033518623454e-06, + "loss": 1.0458, + "step": 1321 + }, + { + "epoch": 0.15896110142487826, + "grad_norm": 2.6480037367756917, + "learning_rate": 3.8281876153631845e-06, + "loss": 1.0548, + "step": 1322 + }, + { + "epoch": 0.15908134431551735, + "grad_norm": 2.988206730331053, + "learning_rate": 3.827871601529416e-06, + "loss": 0.884, + "step": 1323 + }, + { + "epoch": 0.15920158720615643, + "grad_norm": 2.305777288853422, + "learning_rate": 3.827555310408979e-06, + "loss": 1.0419, + "step": 1324 + }, + { + "epoch": 0.1593218300967955, + "grad_norm": 2.9639537415286, + "learning_rate": 3.827238742049854e-06, + "loss": 1.0677, + "step": 1325 + }, + { + "epoch": 0.15944207298743462, + "grad_norm": 1.9185415993505903, + "learning_rate": 3.826921896500066e-06, + "loss": 0.7559, + "step": 1326 + }, + { + "epoch": 0.1595623158780737, + "grad_norm": 6.876370102735521, + "learning_rate": 3.826604773807678e-06, + "loss": 1.0239, + "step": 1327 + }, + { + "epoch": 0.1596825587687128, + "grad_norm": 2.911112665300545, + "learning_rate": 3.826287374020798e-06, + "loss": 0.9724, + "step": 1328 + }, + { + "epoch": 0.1598028016593519, + "grad_norm": 2.824863808648932, + "learning_rate": 3.825969697187575e-06, + "loss": 1.0577, + "step": 1329 + }, + { + "epoch": 0.15992304454999098, + "grad_norm": 1.8813764865467144, + "learning_rate": 3.8256517433562015e-06, + "loss": 0.9348, + "step": 1330 + }, + { + "epoch": 0.16004328744063007, + "grad_norm": 3.1650521696832117, + "learning_rate": 3.82533351257491e-06, + "loss": 1.153, + "step": 1331 + }, + { + "epoch": 0.16016353033126918, + "grad_norm": 5.064522233367182, + "learning_rate": 3.825015004891975e-06, + "loss": 1.1248, + "step": 1332 + }, + { + "epoch": 0.16028377322190826, + "grad_norm": 2.3419626159811604, + "learning_rate": 3.824696220355716e-06, + "loss": 1.0019, + "step": 1333 + }, + { + "epoch": 0.16040401611254734, + "grad_norm": 1.7117651762035395, + "learning_rate": 3.824377159014491e-06, + "loss": 1.0278, + "step": 1334 + }, + { + "epoch": 0.16052425900318643, + "grad_norm": 2.602783466479005, + "learning_rate": 3.824057820916702e-06, + "loss": 1.0901, + "step": 1335 + }, + { + "epoch": 0.16064450189382554, + "grad_norm": 2.9883991691909877, + "learning_rate": 3.8237382061107904e-06, + "loss": 0.9601, + "step": 1336 + }, + { + "epoch": 0.16076474478446462, + "grad_norm": 2.016077271812261, + "learning_rate": 3.823418314645243e-06, + "loss": 1.0244, + "step": 1337 + }, + { + "epoch": 0.1608849876751037, + "grad_norm": 2.2206515907608804, + "learning_rate": 3.823098146568588e-06, + "loss": 0.9931, + "step": 1338 + }, + { + "epoch": 0.1610052305657428, + "grad_norm": 2.3386982707465744, + "learning_rate": 3.822777701929394e-06, + "loss": 0.9516, + "step": 1339 + }, + { + "epoch": 0.1611254734563819, + "grad_norm": 2.1662136362306805, + "learning_rate": 3.8224569807762714e-06, + "loss": 0.9751, + "step": 1340 + }, + { + "epoch": 0.16124571634702098, + "grad_norm": 2.328406195332212, + "learning_rate": 3.822135983157873e-06, + "loss": 1.0062, + "step": 1341 + }, + { + "epoch": 0.16136595923766006, + "grad_norm": 2.3091574559968175, + "learning_rate": 3.821814709122896e-06, + "loss": 1.0826, + "step": 1342 + }, + { + "epoch": 0.16148620212829917, + "grad_norm": 2.871101721927708, + "learning_rate": 3.821493158720076e-06, + "loss": 1.0849, + "step": 1343 + }, + { + "epoch": 0.16160644501893826, + "grad_norm": 3.4918675628140465, + "learning_rate": 3.821171331998191e-06, + "loss": 0.9693, + "step": 1344 + }, + { + "epoch": 0.16172668790957734, + "grad_norm": 0.7093680948601965, + "learning_rate": 3.820849229006064e-06, + "loss": 0.7839, + "step": 1345 + }, + { + "epoch": 0.16184693080021645, + "grad_norm": 2.176819147997458, + "learning_rate": 3.8205268497925564e-06, + "loss": 0.9489, + "step": 1346 + }, + { + "epoch": 0.16196717369085553, + "grad_norm": 5.970614955801408, + "learning_rate": 3.8202041944065725e-06, + "loss": 1.0244, + "step": 1347 + }, + { + "epoch": 0.16208741658149461, + "grad_norm": 2.3040568895270366, + "learning_rate": 3.819881262897061e-06, + "loss": 0.9781, + "step": 1348 + }, + { + "epoch": 0.1622076594721337, + "grad_norm": 2.094138038319145, + "learning_rate": 3.819558055313008e-06, + "loss": 0.9748, + "step": 1349 + }, + { + "epoch": 0.1623279023627728, + "grad_norm": 2.1419792625000422, + "learning_rate": 3.819234571703444e-06, + "loss": 1.0131, + "step": 1350 + }, + { + "epoch": 0.1624481452534119, + "grad_norm": 2.4403995971534296, + "learning_rate": 3.8189108121174435e-06, + "loss": 1.0968, + "step": 1351 + }, + { + "epoch": 0.16256838814405097, + "grad_norm": 2.6646215126418134, + "learning_rate": 3.818586776604118e-06, + "loss": 1.075, + "step": 1352 + }, + { + "epoch": 0.16268863103469008, + "grad_norm": 3.25126813942888, + "learning_rate": 3.818262465212625e-06, + "loss": 0.8578, + "step": 1353 + }, + { + "epoch": 0.16280887392532917, + "grad_norm": 2.4581391424862917, + "learning_rate": 3.817937877992161e-06, + "loss": 1.0086, + "step": 1354 + }, + { + "epoch": 0.16292911681596825, + "grad_norm": 3.8063758065114257, + "learning_rate": 3.817613014991967e-06, + "loss": 1.0965, + "step": 1355 + }, + { + "epoch": 0.16304935970660733, + "grad_norm": 2.1420206685284517, + "learning_rate": 3.817287876261323e-06, + "loss": 1.0033, + "step": 1356 + }, + { + "epoch": 0.16316960259724644, + "grad_norm": 4.992797623245854, + "learning_rate": 3.816962461849553e-06, + "loss": 1.0367, + "step": 1357 + }, + { + "epoch": 0.16328984548788553, + "grad_norm": 2.9154236097381716, + "learning_rate": 3.8166367718060235e-06, + "loss": 1.083, + "step": 1358 + }, + { + "epoch": 0.1634100883785246, + "grad_norm": 2.954321563347193, + "learning_rate": 3.816310806180139e-06, + "loss": 1.0017, + "step": 1359 + }, + { + "epoch": 0.16353033126916372, + "grad_norm": 2.2846686470632407, + "learning_rate": 3.81598456502135e-06, + "loss": 1.0468, + "step": 1360 + }, + { + "epoch": 0.1636505741598028, + "grad_norm": 2.0693172725827655, + "learning_rate": 3.8156580483791455e-06, + "loss": 1.1124, + "step": 1361 + }, + { + "epoch": 0.16377081705044189, + "grad_norm": 2.4607408233040506, + "learning_rate": 3.815331256303059e-06, + "loss": 1.0035, + "step": 1362 + }, + { + "epoch": 0.163891059941081, + "grad_norm": 2.589613407990513, + "learning_rate": 3.815004188842665e-06, + "loss": 1.0083, + "step": 1363 + }, + { + "epoch": 0.16401130283172008, + "grad_norm": 2.2588038396619963, + "learning_rate": 3.814676846047578e-06, + "loss": 1.0345, + "step": 1364 + }, + { + "epoch": 0.16413154572235916, + "grad_norm": 2.0205043905163746, + "learning_rate": 3.8143492279674565e-06, + "loss": 0.9399, + "step": 1365 + }, + { + "epoch": 0.16425178861299825, + "grad_norm": 0.8469252708688886, + "learning_rate": 3.8140213346519997e-06, + "loss": 0.8222, + "step": 1366 + }, + { + "epoch": 0.16437203150363736, + "grad_norm": 1.8632831693435477, + "learning_rate": 3.813693166150948e-06, + "loss": 1.006, + "step": 1367 + }, + { + "epoch": 0.16449227439427644, + "grad_norm": 2.4281798956715313, + "learning_rate": 3.813364722514086e-06, + "loss": 1.0939, + "step": 1368 + }, + { + "epoch": 0.16461251728491552, + "grad_norm": 3.1350669454578544, + "learning_rate": 3.8130360037912368e-06, + "loss": 1.0442, + "step": 1369 + }, + { + "epoch": 0.16473276017555463, + "grad_norm": 2.7718465190744555, + "learning_rate": 3.812707010032268e-06, + "loss": 1.0553, + "step": 1370 + }, + { + "epoch": 0.16485300306619372, + "grad_norm": 3.0781144659956174, + "learning_rate": 3.8123777412870863e-06, + "loss": 1.0328, + "step": 1371 + }, + { + "epoch": 0.1649732459568328, + "grad_norm": 2.481293884965442, + "learning_rate": 3.812048197605643e-06, + "loss": 1.0181, + "step": 1372 + }, + { + "epoch": 0.16509348884747188, + "grad_norm": 2.4745082096897106, + "learning_rate": 3.8117183790379277e-06, + "loss": 1.053, + "step": 1373 + }, + { + "epoch": 0.165213731738111, + "grad_norm": 2.771297176993635, + "learning_rate": 3.811388285633976e-06, + "loss": 1.1779, + "step": 1374 + }, + { + "epoch": 0.16533397462875007, + "grad_norm": 2.3265996454782147, + "learning_rate": 3.811057917443861e-06, + "loss": 0.8576, + "step": 1375 + }, + { + "epoch": 0.16545421751938916, + "grad_norm": 0.8561545283978379, + "learning_rate": 3.8107272745177e-06, + "loss": 0.9213, + "step": 1376 + }, + { + "epoch": 0.16557446041002827, + "grad_norm": 2.568112623306051, + "learning_rate": 3.8103963569056513e-06, + "loss": 1.0248, + "step": 1377 + }, + { + "epoch": 0.16569470330066735, + "grad_norm": 1.7629071137468413, + "learning_rate": 3.8100651646579146e-06, + "loss": 1.1208, + "step": 1378 + }, + { + "epoch": 0.16581494619130643, + "grad_norm": 2.273887668361482, + "learning_rate": 3.8097336978247317e-06, + "loss": 1.1633, + "step": 1379 + }, + { + "epoch": 0.16593518908194552, + "grad_norm": 2.2521139386897318, + "learning_rate": 3.8094019564563854e-06, + "loss": 1.13, + "step": 1380 + }, + { + "epoch": 0.16605543197258463, + "grad_norm": 2.638572662769837, + "learning_rate": 3.809069940603201e-06, + "loss": 0.9867, + "step": 1381 + }, + { + "epoch": 0.1661756748632237, + "grad_norm": 2.458849502512726, + "learning_rate": 3.8087376503155452e-06, + "loss": 1.0173, + "step": 1382 + }, + { + "epoch": 0.1662959177538628, + "grad_norm": 0.8969602630705997, + "learning_rate": 3.808405085643826e-06, + "loss": 0.8002, + "step": 1383 + }, + { + "epoch": 0.1664161606445019, + "grad_norm": 6.228502150933999, + "learning_rate": 3.8080722466384925e-06, + "loss": 1.1284, + "step": 1384 + }, + { + "epoch": 0.166536403535141, + "grad_norm": 2.8486060297206834, + "learning_rate": 3.8077391333500376e-06, + "loss": 0.9487, + "step": 1385 + }, + { + "epoch": 0.16665664642578007, + "grad_norm": 2.183517318234119, + "learning_rate": 3.8074057458289934e-06, + "loss": 1.0048, + "step": 1386 + }, + { + "epoch": 0.16677688931641918, + "grad_norm": 2.6371344796682212, + "learning_rate": 3.807072084125934e-06, + "loss": 1.0649, + "step": 1387 + }, + { + "epoch": 0.16689713220705826, + "grad_norm": 2.5232957735210007, + "learning_rate": 3.806738148291477e-06, + "loss": 1.0433, + "step": 1388 + }, + { + "epoch": 0.16701737509769735, + "grad_norm": 2.29524598775101, + "learning_rate": 3.8064039383762793e-06, + "loss": 0.9512, + "step": 1389 + }, + { + "epoch": 0.16713761798833643, + "grad_norm": 2.892825716847069, + "learning_rate": 3.8060694544310396e-06, + "loss": 1.006, + "step": 1390 + }, + { + "epoch": 0.16725786087897554, + "grad_norm": 2.0458620782867016, + "learning_rate": 3.8057346965065006e-06, + "loss": 1.025, + "step": 1391 + }, + { + "epoch": 0.16737810376961462, + "grad_norm": 1.7765399163118367, + "learning_rate": 3.805399664653443e-06, + "loss": 1.0799, + "step": 1392 + }, + { + "epoch": 0.1674983466602537, + "grad_norm": 2.297939235637525, + "learning_rate": 3.805064358922692e-06, + "loss": 0.9799, + "step": 1393 + }, + { + "epoch": 0.16761858955089282, + "grad_norm": 2.2358984388648913, + "learning_rate": 3.8047287793651136e-06, + "loss": 1.0492, + "step": 1394 + }, + { + "epoch": 0.1677388324415319, + "grad_norm": 2.8905088659103346, + "learning_rate": 3.8043929260316137e-06, + "loss": 1.1262, + "step": 1395 + }, + { + "epoch": 0.16785907533217098, + "grad_norm": 3.3560847293219793, + "learning_rate": 3.8040567989731417e-06, + "loss": 1.0726, + "step": 1396 + }, + { + "epoch": 0.16797931822281006, + "grad_norm": 2.3911121345814843, + "learning_rate": 3.8037203982406876e-06, + "loss": 1.0333, + "step": 1397 + }, + { + "epoch": 0.16809956111344918, + "grad_norm": 2.1214156494921395, + "learning_rate": 3.8033837238852835e-06, + "loss": 0.9689, + "step": 1398 + }, + { + "epoch": 0.16821980400408826, + "grad_norm": 2.675732243221386, + "learning_rate": 3.8030467759580017e-06, + "loss": 0.9325, + "step": 1399 + }, + { + "epoch": 0.16834004689472734, + "grad_norm": 4.564743535503597, + "learning_rate": 3.802709554509958e-06, + "loss": 1.1068, + "step": 1400 + }, + { + "epoch": 0.16846028978536645, + "grad_norm": 2.0991749868597815, + "learning_rate": 3.8023720595923083e-06, + "loss": 1.03, + "step": 1401 + }, + { + "epoch": 0.16858053267600553, + "grad_norm": 4.741921230426167, + "learning_rate": 3.80203429125625e-06, + "loss": 1.1148, + "step": 1402 + }, + { + "epoch": 0.16870077556664462, + "grad_norm": 2.4406006483360203, + "learning_rate": 3.8016962495530225e-06, + "loss": 0.9405, + "step": 1403 + }, + { + "epoch": 0.1688210184572837, + "grad_norm": 4.71513359487352, + "learning_rate": 3.8013579345339063e-06, + "loss": 1.0038, + "step": 1404 + }, + { + "epoch": 0.1689412613479228, + "grad_norm": 4.732296729863581, + "learning_rate": 3.801019346250224e-06, + "loss": 0.9316, + "step": 1405 + }, + { + "epoch": 0.1690615042385619, + "grad_norm": 2.437087684384412, + "learning_rate": 3.8006804847533395e-06, + "loss": 1.0775, + "step": 1406 + }, + { + "epoch": 0.16918174712920098, + "grad_norm": 3.187836722745729, + "learning_rate": 3.8003413500946556e-06, + "loss": 1.0942, + "step": 1407 + }, + { + "epoch": 0.1693019900198401, + "grad_norm": 2.8347518586245934, + "learning_rate": 3.8000019423256216e-06, + "loss": 1.0713, + "step": 1408 + }, + { + "epoch": 0.16942223291047917, + "grad_norm": 2.2138260482916734, + "learning_rate": 3.7996622614977234e-06, + "loss": 1.1173, + "step": 1409 + }, + { + "epoch": 0.16954247580111825, + "grad_norm": 2.0796242948653463, + "learning_rate": 3.799322307662492e-06, + "loss": 1.0293, + "step": 1410 + }, + { + "epoch": 0.16966271869175734, + "grad_norm": 2.3470988444971472, + "learning_rate": 3.798982080871496e-06, + "loss": 1.0731, + "step": 1411 + }, + { + "epoch": 0.16978296158239645, + "grad_norm": 2.5019885737389203, + "learning_rate": 3.798641581176349e-06, + "loss": 0.9179, + "step": 1412 + }, + { + "epoch": 0.16990320447303553, + "grad_norm": 1.8860792256904413, + "learning_rate": 3.7983008086287044e-06, + "loss": 0.9868, + "step": 1413 + }, + { + "epoch": 0.1700234473636746, + "grad_norm": 3.198453296944762, + "learning_rate": 3.797959763280257e-06, + "loss": 1.03, + "step": 1414 + }, + { + "epoch": 0.17014369025431372, + "grad_norm": 2.3143792635563933, + "learning_rate": 3.797618445182743e-06, + "loss": 1.0294, + "step": 1415 + }, + { + "epoch": 0.1702639331449528, + "grad_norm": 2.239162411427214, + "learning_rate": 3.79727685438794e-06, + "loss": 1.0859, + "step": 1416 + }, + { + "epoch": 0.1703841760355919, + "grad_norm": 0.8433961479506115, + "learning_rate": 3.796934990947667e-06, + "loss": 0.8546, + "step": 1417 + }, + { + "epoch": 0.170504418926231, + "grad_norm": 0.8829362833119054, + "learning_rate": 3.7965928549137854e-06, + "loss": 0.8624, + "step": 1418 + }, + { + "epoch": 0.17062466181687008, + "grad_norm": 3.1535891657220776, + "learning_rate": 3.7962504463381953e-06, + "loss": 1.0133, + "step": 1419 + }, + { + "epoch": 0.17074490470750917, + "grad_norm": 1.7862165711934073, + "learning_rate": 3.7959077652728412e-06, + "loss": 1.0242, + "step": 1420 + }, + { + "epoch": 0.17086514759814825, + "grad_norm": 7.593604820339314, + "learning_rate": 3.795564811769707e-06, + "loss": 1.0128, + "step": 1421 + }, + { + "epoch": 0.17098539048878736, + "grad_norm": 8.591326007570622, + "learning_rate": 3.795221585880818e-06, + "loss": 1.0156, + "step": 1422 + }, + { + "epoch": 0.17110563337942644, + "grad_norm": 2.0289356839142165, + "learning_rate": 3.794878087658242e-06, + "loss": 1.1513, + "step": 1423 + }, + { + "epoch": 0.17122587627006552, + "grad_norm": 3.312498848393078, + "learning_rate": 3.7945343171540873e-06, + "loss": 1.0253, + "step": 1424 + }, + { + "epoch": 0.17134611916070464, + "grad_norm": 4.192142475240416, + "learning_rate": 3.7941902744205033e-06, + "loss": 1.027, + "step": 1425 + }, + { + "epoch": 0.17146636205134372, + "grad_norm": 2.338134956817922, + "learning_rate": 3.7938459595096817e-06, + "loss": 1.0743, + "step": 1426 + }, + { + "epoch": 0.1715866049419828, + "grad_norm": 2.064900244273032, + "learning_rate": 3.7935013724738545e-06, + "loss": 1.1026, + "step": 1427 + }, + { + "epoch": 0.17170684783262188, + "grad_norm": 1.9010010667062462, + "learning_rate": 3.7931565133652945e-06, + "loss": 1.0165, + "step": 1428 + }, + { + "epoch": 0.171827090723261, + "grad_norm": 5.592863726598981, + "learning_rate": 3.792811382236317e-06, + "loss": 0.9163, + "step": 1429 + }, + { + "epoch": 0.17194733361390008, + "grad_norm": 2.283616393241635, + "learning_rate": 3.792465979139279e-06, + "loss": 1.0214, + "step": 1430 + }, + { + "epoch": 0.17206757650453916, + "grad_norm": 0.9270447426525652, + "learning_rate": 3.792120304126576e-06, + "loss": 0.8949, + "step": 1431 + }, + { + "epoch": 0.17218781939517827, + "grad_norm": 2.374452628252136, + "learning_rate": 3.791774357250649e-06, + "loss": 1.0748, + "step": 1432 + }, + { + "epoch": 0.17230806228581735, + "grad_norm": 2.110598287145071, + "learning_rate": 3.7914281385639757e-06, + "loss": 1.029, + "step": 1433 + }, + { + "epoch": 0.17242830517645644, + "grad_norm": 2.1117849640903557, + "learning_rate": 3.7910816481190784e-06, + "loss": 1.0324, + "step": 1434 + }, + { + "epoch": 0.17254854806709552, + "grad_norm": 2.323619912842703, + "learning_rate": 3.7907348859685193e-06, + "loss": 0.9878, + "step": 1435 + }, + { + "epoch": 0.17266879095773463, + "grad_norm": 2.345900197767271, + "learning_rate": 3.790387852164902e-06, + "loss": 1.0485, + "step": 1436 + }, + { + "epoch": 0.1727890338483737, + "grad_norm": 2.0167220331373508, + "learning_rate": 3.7900405467608707e-06, + "loss": 1.0019, + "step": 1437 + }, + { + "epoch": 0.1729092767390128, + "grad_norm": 3.3123658350999077, + "learning_rate": 3.7896929698091114e-06, + "loss": 1.0309, + "step": 1438 + }, + { + "epoch": 0.1730295196296519, + "grad_norm": 3.53962578406576, + "learning_rate": 3.7893451213623518e-06, + "loss": 0.9224, + "step": 1439 + }, + { + "epoch": 0.173149762520291, + "grad_norm": 1.9861688390610401, + "learning_rate": 3.7889970014733606e-06, + "loss": 1.0649, + "step": 1440 + }, + { + "epoch": 0.17327000541093007, + "grad_norm": 1.9714629841598847, + "learning_rate": 3.7886486101949463e-06, + "loss": 1.0187, + "step": 1441 + }, + { + "epoch": 0.17339024830156918, + "grad_norm": 3.639267304006366, + "learning_rate": 3.7882999475799594e-06, + "loss": 1.1184, + "step": 1442 + }, + { + "epoch": 0.17351049119220827, + "grad_norm": 2.426298568530333, + "learning_rate": 3.787951013681293e-06, + "loss": 1.0546, + "step": 1443 + }, + { + "epoch": 0.17363073408284735, + "grad_norm": 2.2816657118682686, + "learning_rate": 3.787601808551879e-06, + "loss": 1.0167, + "step": 1444 + }, + { + "epoch": 0.17375097697348643, + "grad_norm": 2.3583549669259742, + "learning_rate": 3.7872523322446926e-06, + "loss": 1.0778, + "step": 1445 + }, + { + "epoch": 0.17387121986412554, + "grad_norm": 2.204175657988335, + "learning_rate": 3.7869025848127478e-06, + "loss": 0.8404, + "step": 1446 + }, + { + "epoch": 0.17399146275476463, + "grad_norm": 2.6231338588809505, + "learning_rate": 3.786552566309102e-06, + "loss": 1.0445, + "step": 1447 + }, + { + "epoch": 0.1741117056454037, + "grad_norm": 2.4562903713956215, + "learning_rate": 3.7862022767868517e-06, + "loss": 1.1045, + "step": 1448 + }, + { + "epoch": 0.17423194853604282, + "grad_norm": 18.629822381934346, + "learning_rate": 3.7858517162991367e-06, + "loss": 1.086, + "step": 1449 + }, + { + "epoch": 0.1743521914266819, + "grad_norm": 4.433897893615263, + "learning_rate": 3.7855008848991363e-06, + "loss": 0.8482, + "step": 1450 + }, + { + "epoch": 0.17447243431732098, + "grad_norm": 2.2149287885322733, + "learning_rate": 3.7851497826400714e-06, + "loss": 1.0101, + "step": 1451 + }, + { + "epoch": 0.17459267720796007, + "grad_norm": 2.5241924845333426, + "learning_rate": 3.7847984095752034e-06, + "loss": 0.9994, + "step": 1452 + }, + { + "epoch": 0.17471292009859918, + "grad_norm": 3.204787106874749, + "learning_rate": 3.784446765757836e-06, + "loss": 1.0393, + "step": 1453 + }, + { + "epoch": 0.17483316298923826, + "grad_norm": 2.6899731811264296, + "learning_rate": 3.7840948512413133e-06, + "loss": 1.0187, + "step": 1454 + }, + { + "epoch": 0.17495340587987734, + "grad_norm": 2.6038517774302945, + "learning_rate": 3.7837426660790196e-06, + "loss": 1.0291, + "step": 1455 + }, + { + "epoch": 0.17507364877051645, + "grad_norm": 4.5297278642912016, + "learning_rate": 3.783390210324382e-06, + "loss": 1.0548, + "step": 1456 + }, + { + "epoch": 0.17519389166115554, + "grad_norm": 2.1025794309292336, + "learning_rate": 3.7830374840308676e-06, + "loss": 0.9648, + "step": 1457 + }, + { + "epoch": 0.17531413455179462, + "grad_norm": 2.7938889914805687, + "learning_rate": 3.7826844872519842e-06, + "loss": 1.0579, + "step": 1458 + }, + { + "epoch": 0.1754343774424337, + "grad_norm": 2.2063809407310737, + "learning_rate": 3.782331220041282e-06, + "loss": 0.9645, + "step": 1459 + }, + { + "epoch": 0.17555462033307281, + "grad_norm": 2.14875774424313, + "learning_rate": 3.7819776824523504e-06, + "loss": 1.0665, + "step": 1460 + }, + { + "epoch": 0.1756748632237119, + "grad_norm": 2.1722289970121027, + "learning_rate": 3.7816238745388213e-06, + "loss": 1.0776, + "step": 1461 + }, + { + "epoch": 0.17579510611435098, + "grad_norm": 2.8267266625958936, + "learning_rate": 3.781269796354367e-06, + "loss": 1.1101, + "step": 1462 + }, + { + "epoch": 0.1759153490049901, + "grad_norm": 1.7554258292926612, + "learning_rate": 3.7809154479527006e-06, + "loss": 1.1016, + "step": 1463 + }, + { + "epoch": 0.17603559189562917, + "grad_norm": 2.2942817217207274, + "learning_rate": 3.780560829387577e-06, + "loss": 1.0825, + "step": 1464 + }, + { + "epoch": 0.17615583478626826, + "grad_norm": 0.8646140533566115, + "learning_rate": 3.7802059407127915e-06, + "loss": 0.8166, + "step": 1465 + }, + { + "epoch": 0.17627607767690734, + "grad_norm": 2.3662929544313416, + "learning_rate": 3.7798507819821797e-06, + "loss": 1.0986, + "step": 1466 + }, + { + "epoch": 0.17639632056754645, + "grad_norm": 4.422933822409297, + "learning_rate": 3.7794953532496197e-06, + "loss": 1.0289, + "step": 1467 + }, + { + "epoch": 0.17651656345818553, + "grad_norm": 0.8616532384710306, + "learning_rate": 3.7791396545690295e-06, + "loss": 0.8178, + "step": 1468 + }, + { + "epoch": 0.17663680634882462, + "grad_norm": 2.2960388322397405, + "learning_rate": 3.7787836859943685e-06, + "loss": 1.0436, + "step": 1469 + }, + { + "epoch": 0.17675704923946373, + "grad_norm": 2.565452177706424, + "learning_rate": 3.7784274475796363e-06, + "loss": 1.0293, + "step": 1470 + }, + { + "epoch": 0.1768772921301028, + "grad_norm": 2.7129739373681545, + "learning_rate": 3.7780709393788745e-06, + "loss": 1.0007, + "step": 1471 + }, + { + "epoch": 0.1769975350207419, + "grad_norm": 2.1395960652964, + "learning_rate": 3.777714161446165e-06, + "loss": 0.9928, + "step": 1472 + }, + { + "epoch": 0.177117777911381, + "grad_norm": 2.6769890280018855, + "learning_rate": 3.7773571138356304e-06, + "loss": 0.9348, + "step": 1473 + }, + { + "epoch": 0.17723802080202009, + "grad_norm": 3.7759708090963127, + "learning_rate": 3.776999796601435e-06, + "loss": 1.1266, + "step": 1474 + }, + { + "epoch": 0.17735826369265917, + "grad_norm": 2.740567938146433, + "learning_rate": 3.776642209797783e-06, + "loss": 0.9722, + "step": 1475 + }, + { + "epoch": 0.17747850658329825, + "grad_norm": 3.4143381945857145, + "learning_rate": 3.7762843534789205e-06, + "loss": 1.0175, + "step": 1476 + }, + { + "epoch": 0.17759874947393736, + "grad_norm": 2.8687563561595573, + "learning_rate": 3.7759262276991343e-06, + "loss": 1.1222, + "step": 1477 + }, + { + "epoch": 0.17771899236457644, + "grad_norm": 2.7587171589081043, + "learning_rate": 3.7755678325127506e-06, + "loss": 1.0434, + "step": 1478 + }, + { + "epoch": 0.17783923525521553, + "grad_norm": 2.2084813578196476, + "learning_rate": 3.7752091679741393e-06, + "loss": 0.9915, + "step": 1479 + }, + { + "epoch": 0.17795947814585464, + "grad_norm": 3.190459429622999, + "learning_rate": 3.774850234137708e-06, + "loss": 1.0143, + "step": 1480 + }, + { + "epoch": 0.17807972103649372, + "grad_norm": 2.441238666123324, + "learning_rate": 3.7744910310579076e-06, + "loss": 1.0665, + "step": 1481 + }, + { + "epoch": 0.1781999639271328, + "grad_norm": 2.437668232125141, + "learning_rate": 3.774131558789229e-06, + "loss": 1.0916, + "step": 1482 + }, + { + "epoch": 0.1783202068177719, + "grad_norm": 11.844997692566853, + "learning_rate": 3.773771817386203e-06, + "loss": 0.9309, + "step": 1483 + }, + { + "epoch": 0.178440449708411, + "grad_norm": 1.7017573193431512, + "learning_rate": 3.773411806903403e-06, + "loss": 1.0306, + "step": 1484 + }, + { + "epoch": 0.17856069259905008, + "grad_norm": 2.0624209302137597, + "learning_rate": 3.7730515273954415e-06, + "loss": 1.1857, + "step": 1485 + }, + { + "epoch": 0.17868093548968916, + "grad_norm": 3.421472673845801, + "learning_rate": 3.772690978916973e-06, + "loss": 1.0896, + "step": 1486 + }, + { + "epoch": 0.17880117838032827, + "grad_norm": 3.8421801330818455, + "learning_rate": 3.772330161522693e-06, + "loss": 1.1065, + "step": 1487 + }, + { + "epoch": 0.17892142127096736, + "grad_norm": 2.1593936726139575, + "learning_rate": 3.7719690752673365e-06, + "loss": 1.0364, + "step": 1488 + }, + { + "epoch": 0.17904166416160644, + "grad_norm": 4.129497041903747, + "learning_rate": 3.7716077202056796e-06, + "loss": 1.0187, + "step": 1489 + }, + { + "epoch": 0.17916190705224552, + "grad_norm": 2.336295008014064, + "learning_rate": 3.7712460963925404e-06, + "loss": 1.1736, + "step": 1490 + }, + { + "epoch": 0.17928214994288463, + "grad_norm": 1.971876840242957, + "learning_rate": 3.7708842038827775e-06, + "loss": 0.9934, + "step": 1491 + }, + { + "epoch": 0.17940239283352372, + "grad_norm": 2.2346056839095874, + "learning_rate": 3.770522042731288e-06, + "loss": 1.0994, + "step": 1492 + }, + { + "epoch": 0.1795226357241628, + "grad_norm": 2.2361187299620977, + "learning_rate": 3.7701596129930122e-06, + "loss": 1.1192, + "step": 1493 + }, + { + "epoch": 0.1796428786148019, + "grad_norm": 2.296410130797756, + "learning_rate": 3.7697969147229315e-06, + "loss": 0.9719, + "step": 1494 + }, + { + "epoch": 0.179763121505441, + "grad_norm": 2.034500924474657, + "learning_rate": 3.7694339479760647e-06, + "loss": 1.0927, + "step": 1495 + }, + { + "epoch": 0.17988336439608008, + "grad_norm": 0.8489793568135761, + "learning_rate": 3.769070712807476e-06, + "loss": 0.8117, + "step": 1496 + }, + { + "epoch": 0.18000360728671919, + "grad_norm": 2.388123244180361, + "learning_rate": 3.768707209272266e-06, + "loss": 1.0256, + "step": 1497 + }, + { + "epoch": 0.18012385017735827, + "grad_norm": 2.7970966272612623, + "learning_rate": 3.768343437425579e-06, + "loss": 0.9998, + "step": 1498 + }, + { + "epoch": 0.18024409306799735, + "grad_norm": 2.6098458784645464, + "learning_rate": 3.7679793973225987e-06, + "loss": 1.0963, + "step": 1499 + }, + { + "epoch": 0.18036433595863643, + "grad_norm": 0.8791287627945769, + "learning_rate": 3.767615089018549e-06, + "loss": 0.8549, + "step": 1500 + }, + { + "epoch": 0.18048457884927555, + "grad_norm": 2.878662099022162, + "learning_rate": 3.7672505125686966e-06, + "loss": 1.0966, + "step": 1501 + }, + { + "epoch": 0.18060482173991463, + "grad_norm": 4.432818885344141, + "learning_rate": 3.7668856680283455e-06, + "loss": 1.0792, + "step": 1502 + }, + { + "epoch": 0.1807250646305537, + "grad_norm": 2.131788406156734, + "learning_rate": 3.7665205554528437e-06, + "loss": 1.0633, + "step": 1503 + }, + { + "epoch": 0.18084530752119282, + "grad_norm": 1.9929417996021768, + "learning_rate": 3.7661551748975782e-06, + "loss": 0.9805, + "step": 1504 + }, + { + "epoch": 0.1809655504118319, + "grad_norm": 0.817245818294587, + "learning_rate": 3.7657895264179772e-06, + "loss": 0.8425, + "step": 1505 + }, + { + "epoch": 0.181085793302471, + "grad_norm": 2.00451543817358, + "learning_rate": 3.765423610069509e-06, + "loss": 0.9836, + "step": 1506 + }, + { + "epoch": 0.18120603619311007, + "grad_norm": 2.003312229208246, + "learning_rate": 3.765057425907683e-06, + "loss": 0.9618, + "step": 1507 + }, + { + "epoch": 0.18132627908374918, + "grad_norm": 3.7503857732073347, + "learning_rate": 3.764690973988048e-06, + "loss": 1.0248, + "step": 1508 + }, + { + "epoch": 0.18144652197438826, + "grad_norm": 3.790515738954363, + "learning_rate": 3.7643242543661967e-06, + "loss": 0.9741, + "step": 1509 + }, + { + "epoch": 0.18156676486502735, + "grad_norm": 0.8409128880239326, + "learning_rate": 3.7639572670977573e-06, + "loss": 0.843, + "step": 1510 + }, + { + "epoch": 0.18168700775566646, + "grad_norm": 1.6224032708533833, + "learning_rate": 3.7635900122384042e-06, + "loss": 1.0041, + "step": 1511 + }, + { + "epoch": 0.18180725064630554, + "grad_norm": 3.4081499592916287, + "learning_rate": 3.7632224898438477e-06, + "loss": 1.1066, + "step": 1512 + }, + { + "epoch": 0.18192749353694462, + "grad_norm": 2.2149601120295777, + "learning_rate": 3.762854699969842e-06, + "loss": 1.0313, + "step": 1513 + }, + { + "epoch": 0.1820477364275837, + "grad_norm": 2.5710062406769083, + "learning_rate": 3.762486642672179e-06, + "loss": 0.97, + "step": 1514 + }, + { + "epoch": 0.18216797931822282, + "grad_norm": 2.1952955577916087, + "learning_rate": 3.7621183180066946e-06, + "loss": 1.1053, + "step": 1515 + }, + { + "epoch": 0.1822882222088619, + "grad_norm": 1.8594192371395377, + "learning_rate": 3.7617497260292625e-06, + "loss": 0.9759, + "step": 1516 + }, + { + "epoch": 0.18240846509950098, + "grad_norm": 3.3335381127080206, + "learning_rate": 3.7613808667957967e-06, + "loss": 1.0233, + "step": 1517 + }, + { + "epoch": 0.1825287079901401, + "grad_norm": 5.6751649000770374, + "learning_rate": 3.7610117403622547e-06, + "loss": 1.1497, + "step": 1518 + }, + { + "epoch": 0.18264895088077918, + "grad_norm": 1.9367197372969511, + "learning_rate": 3.7606423467846313e-06, + "loss": 1.1398, + "step": 1519 + }, + { + "epoch": 0.18276919377141826, + "grad_norm": 1.9089204187765838, + "learning_rate": 3.760272686118964e-06, + "loss": 1.0363, + "step": 1520 + }, + { + "epoch": 0.18288943666205737, + "grad_norm": 2.3585047853868923, + "learning_rate": 3.7599027584213297e-06, + "loss": 1.1622, + "step": 1521 + }, + { + "epoch": 0.18300967955269645, + "grad_norm": 2.086869735139899, + "learning_rate": 3.7595325637478465e-06, + "loss": 1.0204, + "step": 1522 + }, + { + "epoch": 0.18312992244333554, + "grad_norm": 1.9809298307767138, + "learning_rate": 3.7591621021546723e-06, + "loss": 1.0565, + "step": 1523 + }, + { + "epoch": 0.18325016533397462, + "grad_norm": 1.9328530482338524, + "learning_rate": 3.7587913736980062e-06, + "loss": 1.0523, + "step": 1524 + }, + { + "epoch": 0.18337040822461373, + "grad_norm": 4.089467378366124, + "learning_rate": 3.7584203784340865e-06, + "loss": 1.0841, + "step": 1525 + }, + { + "epoch": 0.1834906511152528, + "grad_norm": 2.149307130461407, + "learning_rate": 3.7580491164191938e-06, + "loss": 1.0961, + "step": 1526 + }, + { + "epoch": 0.1836108940058919, + "grad_norm": 0.7512583507765562, + "learning_rate": 3.757677587709648e-06, + "loss": 0.8507, + "step": 1527 + }, + { + "epoch": 0.183731136896531, + "grad_norm": 2.095834406200865, + "learning_rate": 3.7573057923618095e-06, + "loss": 0.9949, + "step": 1528 + }, + { + "epoch": 0.1838513797871701, + "grad_norm": 2.0458205907027285, + "learning_rate": 3.7569337304320793e-06, + "loss": 0.9821, + "step": 1529 + }, + { + "epoch": 0.18397162267780917, + "grad_norm": 0.8403808295861834, + "learning_rate": 3.756561401976899e-06, + "loss": 0.883, + "step": 1530 + }, + { + "epoch": 0.18409186556844825, + "grad_norm": 2.2016560953576154, + "learning_rate": 3.7561888070527514e-06, + "loss": 1.0628, + "step": 1531 + }, + { + "epoch": 0.18421210845908736, + "grad_norm": 2.2230972898320474, + "learning_rate": 3.7558159457161577e-06, + "loss": 1.0382, + "step": 1532 + }, + { + "epoch": 0.18433235134972645, + "grad_norm": 3.1794262528005066, + "learning_rate": 3.755442818023681e-06, + "loss": 1.0198, + "step": 1533 + }, + { + "epoch": 0.18445259424036553, + "grad_norm": 2.7515757121213533, + "learning_rate": 3.7550694240319246e-06, + "loss": 0.9983, + "step": 1534 + }, + { + "epoch": 0.18457283713100464, + "grad_norm": 2.700196460358029, + "learning_rate": 3.7546957637975326e-06, + "loss": 1.0006, + "step": 1535 + }, + { + "epoch": 0.18469308002164372, + "grad_norm": 3.095731774808279, + "learning_rate": 3.7543218373771873e-06, + "loss": 0.9782, + "step": 1536 + }, + { + "epoch": 0.1848133229122828, + "grad_norm": 1.696815141215702, + "learning_rate": 3.753947644827615e-06, + "loss": 1.0192, + "step": 1537 + }, + { + "epoch": 0.1849335658029219, + "grad_norm": 0.9420540507941223, + "learning_rate": 3.753573186205579e-06, + "loss": 0.8112, + "step": 1538 + }, + { + "epoch": 0.185053808693561, + "grad_norm": 2.396631610951771, + "learning_rate": 3.753198461567885e-06, + "loss": 1.0175, + "step": 1539 + }, + { + "epoch": 0.18517405158420008, + "grad_norm": 1.9476345558350738, + "learning_rate": 3.7528234709713783e-06, + "loss": 1.1594, + "step": 1540 + }, + { + "epoch": 0.18529429447483917, + "grad_norm": 2.390306738318991, + "learning_rate": 3.7524482144729447e-06, + "loss": 1.0842, + "step": 1541 + }, + { + "epoch": 0.18541453736547828, + "grad_norm": 2.1207548260600704, + "learning_rate": 3.7520726921295106e-06, + "loss": 1.074, + "step": 1542 + }, + { + "epoch": 0.18553478025611736, + "grad_norm": 2.109941533621032, + "learning_rate": 3.751696903998042e-06, + "loss": 0.9621, + "step": 1543 + }, + { + "epoch": 0.18565502314675644, + "grad_norm": 1.9370036566152804, + "learning_rate": 3.7513208501355456e-06, + "loss": 0.9376, + "step": 1544 + }, + { + "epoch": 0.18577526603739553, + "grad_norm": 2.3714352005227384, + "learning_rate": 3.750944530599069e-06, + "loss": 1.0766, + "step": 1545 + }, + { + "epoch": 0.18589550892803464, + "grad_norm": 3.5568993697926086, + "learning_rate": 3.7505679454456992e-06, + "loss": 1.0448, + "step": 1546 + }, + { + "epoch": 0.18601575181867372, + "grad_norm": 2.481548981225622, + "learning_rate": 3.750191094732564e-06, + "loss": 0.937, + "step": 1547 + }, + { + "epoch": 0.1861359947093128, + "grad_norm": 1.8793044590246901, + "learning_rate": 3.7498139785168313e-06, + "loss": 0.9935, + "step": 1548 + }, + { + "epoch": 0.1862562375999519, + "grad_norm": 1.9058517915637776, + "learning_rate": 3.749436596855709e-06, + "loss": 1.0091, + "step": 1549 + }, + { + "epoch": 0.186376480490591, + "grad_norm": 2.412959664783494, + "learning_rate": 3.749058949806446e-06, + "loss": 1.1435, + "step": 1550 + }, + { + "epoch": 0.18649672338123008, + "grad_norm": 1.8228108039271251, + "learning_rate": 3.748681037426331e-06, + "loss": 1.0814, + "step": 1551 + }, + { + "epoch": 0.1866169662718692, + "grad_norm": 2.1140884296324183, + "learning_rate": 3.7483028597726936e-06, + "loss": 1.1557, + "step": 1552 + }, + { + "epoch": 0.18673720916250827, + "grad_norm": 2.423209136690765, + "learning_rate": 3.7479244169029017e-06, + "loss": 0.8644, + "step": 1553 + }, + { + "epoch": 0.18685745205314735, + "grad_norm": 2.7401712944617063, + "learning_rate": 3.7475457088743658e-06, + "loss": 0.9682, + "step": 1554 + }, + { + "epoch": 0.18697769494378644, + "grad_norm": 2.9087575943687454, + "learning_rate": 3.7471667357445348e-06, + "loss": 0.9851, + "step": 1555 + }, + { + "epoch": 0.18709793783442555, + "grad_norm": 2.140603169796306, + "learning_rate": 3.7467874975709e-06, + "loss": 0.9645, + "step": 1556 + }, + { + "epoch": 0.18721818072506463, + "grad_norm": 2.3738565201886956, + "learning_rate": 3.7464079944109904e-06, + "loss": 1.0208, + "step": 1557 + }, + { + "epoch": 0.18733842361570371, + "grad_norm": 2.3485220583618243, + "learning_rate": 3.746028226322376e-06, + "loss": 1.0151, + "step": 1558 + }, + { + "epoch": 0.18745866650634282, + "grad_norm": 1.809685396177755, + "learning_rate": 3.745648193362669e-06, + "loss": 0.9917, + "step": 1559 + }, + { + "epoch": 0.1875789093969819, + "grad_norm": 2.0981285657446147, + "learning_rate": 3.745267895589518e-06, + "loss": 0.9642, + "step": 1560 + }, + { + "epoch": 0.187699152287621, + "grad_norm": 2.1042169873354934, + "learning_rate": 3.7448873330606154e-06, + "loss": 1.0609, + "step": 1561 + }, + { + "epoch": 0.18781939517826007, + "grad_norm": 2.5207079601374476, + "learning_rate": 3.7445065058336914e-06, + "loss": 1.1106, + "step": 1562 + }, + { + "epoch": 0.18793963806889918, + "grad_norm": 2.177929678742377, + "learning_rate": 3.7441254139665176e-06, + "loss": 1.1014, + "step": 1563 + }, + { + "epoch": 0.18805988095953827, + "grad_norm": 1.796225123435612, + "learning_rate": 3.743744057516905e-06, + "loss": 1.063, + "step": 1564 + }, + { + "epoch": 0.18818012385017735, + "grad_norm": 3.0351196443959734, + "learning_rate": 3.743362436542706e-06, + "loss": 1.1152, + "step": 1565 + }, + { + "epoch": 0.18830036674081646, + "grad_norm": 1.8913916932735095, + "learning_rate": 3.7429805511018115e-06, + "loss": 1.0049, + "step": 1566 + }, + { + "epoch": 0.18842060963145554, + "grad_norm": 1.900877965558944, + "learning_rate": 3.7425984012521524e-06, + "loss": 1.018, + "step": 1567 + }, + { + "epoch": 0.18854085252209463, + "grad_norm": 1.2511952407851281, + "learning_rate": 3.7422159870517025e-06, + "loss": 0.8424, + "step": 1568 + }, + { + "epoch": 0.1886610954127337, + "grad_norm": 2.093173844504238, + "learning_rate": 3.7418333085584717e-06, + "loss": 1.0256, + "step": 1569 + }, + { + "epoch": 0.18878133830337282, + "grad_norm": 2.0386417583310306, + "learning_rate": 3.7414503658305128e-06, + "loss": 1.1479, + "step": 1570 + }, + { + "epoch": 0.1889015811940119, + "grad_norm": 2.507376849876652, + "learning_rate": 3.7410671589259185e-06, + "loss": 1.0185, + "step": 1571 + }, + { + "epoch": 0.18902182408465099, + "grad_norm": 1.9382854223302535, + "learning_rate": 3.7406836879028205e-06, + "loss": 1.0351, + "step": 1572 + }, + { + "epoch": 0.1891420669752901, + "grad_norm": 4.804931038987834, + "learning_rate": 3.7402999528193907e-06, + "loss": 1.0112, + "step": 1573 + }, + { + "epoch": 0.18926230986592918, + "grad_norm": 2.6053422842831853, + "learning_rate": 3.739915953733842e-06, + "loss": 1.0954, + "step": 1574 + }, + { + "epoch": 0.18938255275656826, + "grad_norm": 1.6394379453976575, + "learning_rate": 3.7395316907044264e-06, + "loss": 1.0576, + "step": 1575 + }, + { + "epoch": 0.18950279564720737, + "grad_norm": 1.5938759548306158, + "learning_rate": 3.7391471637894364e-06, + "loss": 1.0345, + "step": 1576 + }, + { + "epoch": 0.18962303853784646, + "grad_norm": 1.9288707657273687, + "learning_rate": 3.738762373047205e-06, + "loss": 1.0889, + "step": 1577 + }, + { + "epoch": 0.18974328142848554, + "grad_norm": 1.6835996269247413, + "learning_rate": 3.738377318536103e-06, + "loss": 1.0734, + "step": 1578 + }, + { + "epoch": 0.18986352431912462, + "grad_norm": 2.305831263295044, + "learning_rate": 3.7379920003145447e-06, + "loss": 0.9578, + "step": 1579 + }, + { + "epoch": 0.18998376720976373, + "grad_norm": 1.95082658952571, + "learning_rate": 3.7376064184409817e-06, + "loss": 1.0777, + "step": 1580 + }, + { + "epoch": 0.19010401010040281, + "grad_norm": 1.469021548877236, + "learning_rate": 3.7372205729739063e-06, + "loss": 1.1112, + "step": 1581 + }, + { + "epoch": 0.1902242529910419, + "grad_norm": 4.439226056596733, + "learning_rate": 3.7368344639718514e-06, + "loss": 0.9539, + "step": 1582 + }, + { + "epoch": 0.190344495881681, + "grad_norm": 1.9565240808141213, + "learning_rate": 3.7364480914933895e-06, + "loss": 1.0438, + "step": 1583 + }, + { + "epoch": 0.1904647387723201, + "grad_norm": 1.8695049508807935, + "learning_rate": 3.7360614555971325e-06, + "loss": 1.0488, + "step": 1584 + }, + { + "epoch": 0.19058498166295917, + "grad_norm": 2.0437520205053206, + "learning_rate": 3.735674556341733e-06, + "loss": 1.092, + "step": 1585 + }, + { + "epoch": 0.19070522455359826, + "grad_norm": 3.3724947921000115, + "learning_rate": 3.7352873937858835e-06, + "loss": 1.0672, + "step": 1586 + }, + { + "epoch": 0.19082546744423737, + "grad_norm": 2.1317528408349684, + "learning_rate": 3.734899967988316e-06, + "loss": 0.959, + "step": 1587 + }, + { + "epoch": 0.19094571033487645, + "grad_norm": 2.6221024097488437, + "learning_rate": 3.7345122790078026e-06, + "loss": 1.0803, + "step": 1588 + }, + { + "epoch": 0.19106595322551553, + "grad_norm": 2.490934912187665, + "learning_rate": 3.7341243269031556e-06, + "loss": 1.1678, + "step": 1589 + }, + { + "epoch": 0.19118619611615464, + "grad_norm": 1.9829213382318303, + "learning_rate": 3.7337361117332275e-06, + "loss": 1.0108, + "step": 1590 + }, + { + "epoch": 0.19130643900679373, + "grad_norm": 2.0222744342272394, + "learning_rate": 3.7333476335569087e-06, + "loss": 1.0086, + "step": 1591 + }, + { + "epoch": 0.1914266818974328, + "grad_norm": 3.3790271891741597, + "learning_rate": 3.7329588924331325e-06, + "loss": 0.9084, + "step": 1592 + }, + { + "epoch": 0.1915469247880719, + "grad_norm": 2.20266026133575, + "learning_rate": 3.732569888420871e-06, + "loss": 1.0656, + "step": 1593 + }, + { + "epoch": 0.191667167678711, + "grad_norm": 6.257439115245618, + "learning_rate": 3.732180621579134e-06, + "loss": 1.059, + "step": 1594 + }, + { + "epoch": 0.1917874105693501, + "grad_norm": 2.466853415233121, + "learning_rate": 3.7317910919669745e-06, + "loss": 1.0539, + "step": 1595 + }, + { + "epoch": 0.19190765345998917, + "grad_norm": 3.2712550934113094, + "learning_rate": 3.7314012996434826e-06, + "loss": 1.0013, + "step": 1596 + }, + { + "epoch": 0.19202789635062828, + "grad_norm": 2.1935783949460683, + "learning_rate": 3.7310112446677907e-06, + "loss": 1.0508, + "step": 1597 + }, + { + "epoch": 0.19214813924126736, + "grad_norm": 2.2414901659186266, + "learning_rate": 3.7306209270990695e-06, + "loss": 0.9292, + "step": 1598 + }, + { + "epoch": 0.19226838213190645, + "grad_norm": 1.9730210372994248, + "learning_rate": 3.7302303469965292e-06, + "loss": 1.1064, + "step": 1599 + }, + { + "epoch": 0.19238862502254553, + "grad_norm": 1.9411955014824205, + "learning_rate": 3.7298395044194206e-06, + "loss": 0.9458, + "step": 1600 + }, + { + "epoch": 0.19250886791318464, + "grad_norm": 4.5624420998439765, + "learning_rate": 3.7294483994270356e-06, + "loss": 1.1809, + "step": 1601 + }, + { + "epoch": 0.19262911080382372, + "grad_norm": 3.491537080482922, + "learning_rate": 3.7290570320787033e-06, + "loss": 1.0197, + "step": 1602 + }, + { + "epoch": 0.1927493536944628, + "grad_norm": 1.8214283416918868, + "learning_rate": 3.728665402433793e-06, + "loss": 0.9528, + "step": 1603 + }, + { + "epoch": 0.19286959658510192, + "grad_norm": 2.3681341613769518, + "learning_rate": 3.7282735105517164e-06, + "loss": 1.0988, + "step": 1604 + }, + { + "epoch": 0.192989839475741, + "grad_norm": 2.104308762489337, + "learning_rate": 3.727881356491922e-06, + "loss": 0.9107, + "step": 1605 + }, + { + "epoch": 0.19311008236638008, + "grad_norm": 2.3040565790938223, + "learning_rate": 3.7274889403139002e-06, + "loss": 0.9944, + "step": 1606 + }, + { + "epoch": 0.1932303252570192, + "grad_norm": 2.305620735145899, + "learning_rate": 3.727096262077179e-06, + "loss": 1.0209, + "step": 1607 + }, + { + "epoch": 0.19335056814765827, + "grad_norm": 1.8049049411585787, + "learning_rate": 3.7267033218413285e-06, + "loss": 1.091, + "step": 1608 + }, + { + "epoch": 0.19347081103829736, + "grad_norm": 2.220580299758552, + "learning_rate": 3.726310119665957e-06, + "loss": 1.0524, + "step": 1609 + }, + { + "epoch": 0.19359105392893644, + "grad_norm": 1.9476366980864666, + "learning_rate": 3.725916655610713e-06, + "loss": 1.0916, + "step": 1610 + }, + { + "epoch": 0.19371129681957555, + "grad_norm": 3.206900999372352, + "learning_rate": 3.725522929735284e-06, + "loss": 0.9938, + "step": 1611 + }, + { + "epoch": 0.19383153971021463, + "grad_norm": 2.211686617376306, + "learning_rate": 3.725128942099399e-06, + "loss": 0.9791, + "step": 1612 + }, + { + "epoch": 0.19395178260085372, + "grad_norm": 3.1247267031373025, + "learning_rate": 3.7247346927628245e-06, + "loss": 1.0373, + "step": 1613 + }, + { + "epoch": 0.19407202549149283, + "grad_norm": 2.006182651042946, + "learning_rate": 3.7243401817853694e-06, + "loss": 1.0305, + "step": 1614 + }, + { + "epoch": 0.1941922683821319, + "grad_norm": 2.1778963995219747, + "learning_rate": 3.723945409226879e-06, + "loss": 0.9557, + "step": 1615 + }, + { + "epoch": 0.194312511272771, + "grad_norm": 2.4422140265236534, + "learning_rate": 3.723550375147241e-06, + "loss": 1.0413, + "step": 1616 + }, + { + "epoch": 0.19443275416341008, + "grad_norm": 2.7235333686802776, + "learning_rate": 3.7231550796063816e-06, + "loss": 1.0336, + "step": 1617 + }, + { + "epoch": 0.1945529970540492, + "grad_norm": 2.0855259864448388, + "learning_rate": 3.722759522664266e-06, + "loss": 0.8934, + "step": 1618 + }, + { + "epoch": 0.19467323994468827, + "grad_norm": 2.4074191931357265, + "learning_rate": 3.7223637043809016e-06, + "loss": 1.0573, + "step": 1619 + }, + { + "epoch": 0.19479348283532735, + "grad_norm": 6.560879171120809, + "learning_rate": 3.7219676248163322e-06, + "loss": 1.1038, + "step": 1620 + }, + { + "epoch": 0.19491372572596646, + "grad_norm": 1.8945397445645582, + "learning_rate": 3.721571284030643e-06, + "loss": 1.1735, + "step": 1621 + }, + { + "epoch": 0.19503396861660555, + "grad_norm": 2.2631897565146675, + "learning_rate": 3.7211746820839587e-06, + "loss": 1.0316, + "step": 1622 + }, + { + "epoch": 0.19515421150724463, + "grad_norm": 1.7642141461128598, + "learning_rate": 3.7207778190364437e-06, + "loss": 1.0461, + "step": 1623 + }, + { + "epoch": 0.1952744543978837, + "grad_norm": 1.6931984372862399, + "learning_rate": 3.720380694948302e-06, + "loss": 0.9776, + "step": 1624 + }, + { + "epoch": 0.19539469728852282, + "grad_norm": 1.0364302946096842, + "learning_rate": 3.719983309879777e-06, + "loss": 0.9504, + "step": 1625 + }, + { + "epoch": 0.1955149401791619, + "grad_norm": 1.670359842969437, + "learning_rate": 3.719585663891151e-06, + "loss": 1.0126, + "step": 1626 + }, + { + "epoch": 0.195635183069801, + "grad_norm": 2.2859907664248533, + "learning_rate": 3.719187757042747e-06, + "loss": 1.0331, + "step": 1627 + }, + { + "epoch": 0.1957554259604401, + "grad_norm": 0.7261774006469688, + "learning_rate": 3.7187895893949275e-06, + "loss": 0.7876, + "step": 1628 + }, + { + "epoch": 0.19587566885107918, + "grad_norm": 2.8092930630116713, + "learning_rate": 3.7183911610080937e-06, + "loss": 1.0009, + "step": 1629 + }, + { + "epoch": 0.19599591174171827, + "grad_norm": 2.22450927294811, + "learning_rate": 3.7179924719426872e-06, + "loss": 0.9866, + "step": 1630 + }, + { + "epoch": 0.19611615463235738, + "grad_norm": 2.593297712119, + "learning_rate": 3.7175935222591885e-06, + "loss": 1.0016, + "step": 1631 + }, + { + "epoch": 0.19623639752299646, + "grad_norm": 1.9083701053835715, + "learning_rate": 3.717194312018118e-06, + "loss": 0.9832, + "step": 1632 + }, + { + "epoch": 0.19635664041363554, + "grad_norm": 3.7364001030377447, + "learning_rate": 3.716794841280036e-06, + "loss": 1.003, + "step": 1633 + }, + { + "epoch": 0.19647688330427462, + "grad_norm": 2.4062144289236165, + "learning_rate": 3.7163951101055407e-06, + "loss": 1.0154, + "step": 1634 + }, + { + "epoch": 0.19659712619491373, + "grad_norm": 5.336245615875381, + "learning_rate": 3.715995118555273e-06, + "loss": 1.0289, + "step": 1635 + }, + { + "epoch": 0.19671736908555282, + "grad_norm": 3.754196489210714, + "learning_rate": 3.71559486668991e-06, + "loss": 1.1001, + "step": 1636 + }, + { + "epoch": 0.1968376119761919, + "grad_norm": 1.8444410419177393, + "learning_rate": 3.715194354570169e-06, + "loss": 1.0119, + "step": 1637 + }, + { + "epoch": 0.196957854866831, + "grad_norm": 1.9496437480925377, + "learning_rate": 3.714793582256809e-06, + "loss": 1.0714, + "step": 1638 + }, + { + "epoch": 0.1970780977574701, + "grad_norm": 2.8801401048595894, + "learning_rate": 3.7143925498106253e-06, + "loss": 1.0913, + "step": 1639 + }, + { + "epoch": 0.19719834064810918, + "grad_norm": 1.8526263178746114, + "learning_rate": 3.7139912572924558e-06, + "loss": 1.0296, + "step": 1640 + }, + { + "epoch": 0.19731858353874826, + "grad_norm": 2.3673673756560896, + "learning_rate": 3.7135897047631744e-06, + "loss": 1.0461, + "step": 1641 + }, + { + "epoch": 0.19743882642938737, + "grad_norm": 3.14798205089178, + "learning_rate": 3.713187892283698e-06, + "loss": 1.0007, + "step": 1642 + }, + { + "epoch": 0.19755906932002645, + "grad_norm": 2.2015414126640462, + "learning_rate": 3.71278581991498e-06, + "loss": 1.1153, + "step": 1643 + }, + { + "epoch": 0.19767931221066554, + "grad_norm": 1.9308355213801016, + "learning_rate": 3.712383487718015e-06, + "loss": 1.0277, + "step": 1644 + }, + { + "epoch": 0.19779955510130465, + "grad_norm": 1.9490596117637706, + "learning_rate": 3.7119808957538365e-06, + "loss": 1.108, + "step": 1645 + }, + { + "epoch": 0.19791979799194373, + "grad_norm": 2.1326406779962, + "learning_rate": 3.711578044083517e-06, + "loss": 1.0395, + "step": 1646 + }, + { + "epoch": 0.1980400408825828, + "grad_norm": 5.50095047107362, + "learning_rate": 3.7111749327681698e-06, + "loss": 0.9885, + "step": 1647 + }, + { + "epoch": 0.1981602837732219, + "grad_norm": 2.279008364947887, + "learning_rate": 3.7107715618689455e-06, + "loss": 1.1022, + "step": 1648 + }, + { + "epoch": 0.198280526663861, + "grad_norm": 1.4757798153984887, + "learning_rate": 3.710367931447035e-06, + "loss": 1.0736, + "step": 1649 + }, + { + "epoch": 0.1984007695545001, + "grad_norm": 2.2575870388145582, + "learning_rate": 3.70996404156367e-06, + "loss": 1.1046, + "step": 1650 + }, + { + "epoch": 0.19852101244513917, + "grad_norm": 1.7563016374809417, + "learning_rate": 3.7095598922801187e-06, + "loss": 0.9655, + "step": 1651 + }, + { + "epoch": 0.19864125533577828, + "grad_norm": 2.221513129058788, + "learning_rate": 3.7091554836576914e-06, + "loss": 1.0002, + "step": 1652 + }, + { + "epoch": 0.19876149822641737, + "grad_norm": 2.5285253106469057, + "learning_rate": 3.708750815757736e-06, + "loss": 1.0692, + "step": 1653 + }, + { + "epoch": 0.19888174111705645, + "grad_norm": 8.27887366406251, + "learning_rate": 3.7083458886416407e-06, + "loss": 0.972, + "step": 1654 + }, + { + "epoch": 0.19900198400769553, + "grad_norm": 2.078634787536294, + "learning_rate": 3.707940702370832e-06, + "loss": 1.1203, + "step": 1655 + }, + { + "epoch": 0.19912222689833464, + "grad_norm": 0.768301431306645, + "learning_rate": 3.707535257006777e-06, + "loss": 0.8203, + "step": 1656 + }, + { + "epoch": 0.19924246978897373, + "grad_norm": 2.8021804528027063, + "learning_rate": 3.707129552610981e-06, + "loss": 1.1222, + "step": 1657 + }, + { + "epoch": 0.1993627126796128, + "grad_norm": 2.1322470114435754, + "learning_rate": 3.70672358924499e-06, + "loss": 0.977, + "step": 1658 + }, + { + "epoch": 0.19948295557025192, + "grad_norm": 2.0786716057818846, + "learning_rate": 3.706317366970386e-06, + "loss": 1.0209, + "step": 1659 + }, + { + "epoch": 0.199603198460891, + "grad_norm": 2.4644964732204655, + "learning_rate": 3.705910885848795e-06, + "loss": 1.0757, + "step": 1660 + }, + { + "epoch": 0.19972344135153008, + "grad_norm": 2.070838947659512, + "learning_rate": 3.705504145941879e-06, + "loss": 1.0875, + "step": 1661 + }, + { + "epoch": 0.1998436842421692, + "grad_norm": 2.2868343614677364, + "learning_rate": 3.7050971473113403e-06, + "loss": 1.0251, + "step": 1662 + }, + { + "epoch": 0.19996392713280828, + "grad_norm": 1.8146840452618211, + "learning_rate": 3.7046898900189196e-06, + "loss": 1.0381, + "step": 1663 + }, + { + "epoch": 0.20008417002344736, + "grad_norm": 1.868769466337033, + "learning_rate": 3.704282374126398e-06, + "loss": 1.0656, + "step": 1664 + }, + { + "epoch": 0.20020441291408644, + "grad_norm": 1.9466224763810156, + "learning_rate": 3.7038745996955954e-06, + "loss": 1.1107, + "step": 1665 + }, + { + "epoch": 0.20032465580472555, + "grad_norm": 2.8382824327344083, + "learning_rate": 3.703466566788371e-06, + "loss": 0.9521, + "step": 1666 + }, + { + "epoch": 0.20044489869536464, + "grad_norm": 2.3003915868035856, + "learning_rate": 3.703058275466622e-06, + "loss": 0.9819, + "step": 1667 + }, + { + "epoch": 0.20056514158600372, + "grad_norm": 2.6212722103576835, + "learning_rate": 3.7026497257922877e-06, + "loss": 1.0125, + "step": 1668 + }, + { + "epoch": 0.20068538447664283, + "grad_norm": 2.4410560295678887, + "learning_rate": 3.7022409178273436e-06, + "loss": 1.0947, + "step": 1669 + }, + { + "epoch": 0.2008056273672819, + "grad_norm": 1.8978160578849679, + "learning_rate": 3.7018318516338054e-06, + "loss": 1.0256, + "step": 1670 + }, + { + "epoch": 0.200925870257921, + "grad_norm": 4.895079123117626, + "learning_rate": 3.7014225272737284e-06, + "loss": 1.0628, + "step": 1671 + }, + { + "epoch": 0.20104611314856008, + "grad_norm": 2.640969699730578, + "learning_rate": 3.701012944809207e-06, + "loss": 0.9818, + "step": 1672 + }, + { + "epoch": 0.2011663560391992, + "grad_norm": 2.3470036619959944, + "learning_rate": 3.700603104302374e-06, + "loss": 1.0299, + "step": 1673 + }, + { + "epoch": 0.20128659892983827, + "grad_norm": 0.9111314060081755, + "learning_rate": 3.7001930058154027e-06, + "loss": 0.7984, + "step": 1674 + }, + { + "epoch": 0.20140684182047736, + "grad_norm": 3.615812827253839, + "learning_rate": 3.6997826494105037e-06, + "loss": 1.0383, + "step": 1675 + }, + { + "epoch": 0.20152708471111647, + "grad_norm": 2.8091184844806873, + "learning_rate": 3.6993720351499286e-06, + "loss": 0.9305, + "step": 1676 + }, + { + "epoch": 0.20164732760175555, + "grad_norm": 1.9178414545333413, + "learning_rate": 3.6989611630959666e-06, + "loss": 1.01, + "step": 1677 + }, + { + "epoch": 0.20176757049239463, + "grad_norm": 0.7773373090774301, + "learning_rate": 3.6985500333109474e-06, + "loss": 0.8216, + "step": 1678 + }, + { + "epoch": 0.20188781338303372, + "grad_norm": 2.6184034568903662, + "learning_rate": 3.6981386458572385e-06, + "loss": 1.0032, + "step": 1679 + }, + { + "epoch": 0.20200805627367283, + "grad_norm": 2.414702790415425, + "learning_rate": 3.6977270007972468e-06, + "loss": 0.9947, + "step": 1680 + }, + { + "epoch": 0.2021282991643119, + "grad_norm": 3.039088545739129, + "learning_rate": 3.6973150981934196e-06, + "loss": 0.9601, + "step": 1681 + }, + { + "epoch": 0.202248542054951, + "grad_norm": 2.539718214609475, + "learning_rate": 3.6969029381082415e-06, + "loss": 1.0812, + "step": 1682 + }, + { + "epoch": 0.2023687849455901, + "grad_norm": 1.9852514659231029, + "learning_rate": 3.696490520604237e-06, + "loss": 1.0322, + "step": 1683 + }, + { + "epoch": 0.20248902783622919, + "grad_norm": 1.879600983338587, + "learning_rate": 3.696077845743968e-06, + "loss": 1.0447, + "step": 1684 + }, + { + "epoch": 0.20260927072686827, + "grad_norm": 2.515090032524502, + "learning_rate": 3.69566491359004e-06, + "loss": 0.9745, + "step": 1685 + }, + { + "epoch": 0.20272951361750738, + "grad_norm": 19.210911288869877, + "learning_rate": 3.695251724205092e-06, + "loss": 0.9354, + "step": 1686 + }, + { + "epoch": 0.20284975650814646, + "grad_norm": 1.6257184714460493, + "learning_rate": 3.6948382776518054e-06, + "loss": 1.1035, + "step": 1687 + }, + { + "epoch": 0.20296999939878554, + "grad_norm": 3.0762034968612793, + "learning_rate": 3.6944245739929e-06, + "loss": 1.0371, + "step": 1688 + }, + { + "epoch": 0.20309024228942463, + "grad_norm": 2.202754320925998, + "learning_rate": 3.6940106132911332e-06, + "loss": 0.9568, + "step": 1689 + }, + { + "epoch": 0.20321048518006374, + "grad_norm": 2.1557217379326263, + "learning_rate": 3.6935963956093037e-06, + "loss": 1.1256, + "step": 1690 + }, + { + "epoch": 0.20333072807070282, + "grad_norm": 2.0469975325621252, + "learning_rate": 3.6931819210102474e-06, + "loss": 0.9313, + "step": 1691 + }, + { + "epoch": 0.2034509709613419, + "grad_norm": 1.7742752583557406, + "learning_rate": 3.6927671895568402e-06, + "loss": 1.082, + "step": 1692 + }, + { + "epoch": 0.20357121385198101, + "grad_norm": 3.3737363922308297, + "learning_rate": 3.692352201311996e-06, + "loss": 1.1087, + "step": 1693 + }, + { + "epoch": 0.2036914567426201, + "grad_norm": 1.9002989132305592, + "learning_rate": 3.6919369563386687e-06, + "loss": 1.0042, + "step": 1694 + }, + { + "epoch": 0.20381169963325918, + "grad_norm": 2.5409597966880564, + "learning_rate": 3.69152145469985e-06, + "loss": 1.0344, + "step": 1695 + }, + { + "epoch": 0.20393194252389826, + "grad_norm": 2.185316467412475, + "learning_rate": 3.691105696458572e-06, + "loss": 1.06, + "step": 1696 + }, + { + "epoch": 0.20405218541453737, + "grad_norm": 2.4436805812800486, + "learning_rate": 3.690689681677904e-06, + "loss": 0.9162, + "step": 1697 + }, + { + "epoch": 0.20417242830517646, + "grad_norm": 1.8408343003447332, + "learning_rate": 3.690273410420956e-06, + "loss": 1.129, + "step": 1698 + }, + { + "epoch": 0.20429267119581554, + "grad_norm": 2.364560725122222, + "learning_rate": 3.689856882750875e-06, + "loss": 1.0035, + "step": 1699 + }, + { + "epoch": 0.20441291408645465, + "grad_norm": 1.8165982708902686, + "learning_rate": 3.6894400987308486e-06, + "loss": 1.0245, + "step": 1700 + }, + { + "epoch": 0.20453315697709373, + "grad_norm": 2.7947661904960484, + "learning_rate": 3.6890230584241024e-06, + "loss": 1.0879, + "step": 1701 + }, + { + "epoch": 0.20465339986773282, + "grad_norm": 1.0823881476498398, + "learning_rate": 3.6886057618939016e-06, + "loss": 0.9015, + "step": 1702 + }, + { + "epoch": 0.2047736427583719, + "grad_norm": 2.3127200047769736, + "learning_rate": 3.6881882092035492e-06, + "loss": 0.9309, + "step": 1703 + }, + { + "epoch": 0.204893885649011, + "grad_norm": 0.9198066655480891, + "learning_rate": 3.6877704004163873e-06, + "loss": 0.8504, + "step": 1704 + }, + { + "epoch": 0.2050141285396501, + "grad_norm": 1.9243361678484476, + "learning_rate": 3.6873523355957984e-06, + "loss": 1.0157, + "step": 1705 + }, + { + "epoch": 0.20513437143028918, + "grad_norm": 0.9973999435034675, + "learning_rate": 3.686934014805201e-06, + "loss": 0.8782, + "step": 1706 + }, + { + "epoch": 0.20525461432092829, + "grad_norm": 2.0041391932560035, + "learning_rate": 3.6865154381080552e-06, + "loss": 1.0462, + "step": 1707 + }, + { + "epoch": 0.20537485721156737, + "grad_norm": 2.035580288867712, + "learning_rate": 3.6860966055678585e-06, + "loss": 1.0665, + "step": 1708 + }, + { + "epoch": 0.20549510010220645, + "grad_norm": 2.136085765188335, + "learning_rate": 3.685677517248147e-06, + "loss": 1.1053, + "step": 1709 + }, + { + "epoch": 0.20561534299284553, + "grad_norm": 2.1062642082844443, + "learning_rate": 3.6852581732124967e-06, + "loss": 1.0424, + "step": 1710 + }, + { + "epoch": 0.20573558588348465, + "grad_norm": 3.036050672085642, + "learning_rate": 3.6848385735245213e-06, + "loss": 1.001, + "step": 1711 + }, + { + "epoch": 0.20585582877412373, + "grad_norm": 1.9862666687887143, + "learning_rate": 3.6844187182478734e-06, + "loss": 1.1002, + "step": 1712 + }, + { + "epoch": 0.2059760716647628, + "grad_norm": 10.692932889611637, + "learning_rate": 3.683998607446246e-06, + "loss": 0.9883, + "step": 1713 + }, + { + "epoch": 0.20609631455540192, + "grad_norm": 1.9338127243432937, + "learning_rate": 3.6835782411833686e-06, + "loss": 0.99, + "step": 1714 + }, + { + "epoch": 0.206216557446041, + "grad_norm": 2.7862322469002816, + "learning_rate": 3.68315761952301e-06, + "loss": 0.9858, + "step": 1715 + }, + { + "epoch": 0.2063368003366801, + "grad_norm": 2.0355668193782206, + "learning_rate": 3.6827367425289797e-06, + "loss": 1.0741, + "step": 1716 + }, + { + "epoch": 0.2064570432273192, + "grad_norm": 3.611659924463147, + "learning_rate": 3.6823156102651225e-06, + "loss": 0.9666, + "step": 1717 + }, + { + "epoch": 0.20657728611795828, + "grad_norm": 1.9262587630402441, + "learning_rate": 3.6818942227953257e-06, + "loss": 0.9496, + "step": 1718 + }, + { + "epoch": 0.20669752900859736, + "grad_norm": 2.0865385068388362, + "learning_rate": 3.681472580183512e-06, + "loss": 0.9319, + "step": 1719 + }, + { + "epoch": 0.20681777189923645, + "grad_norm": 2.074040119647572, + "learning_rate": 3.6810506824936455e-06, + "loss": 1.1014, + "step": 1720 + }, + { + "epoch": 0.20693801478987556, + "grad_norm": 1.0340041749194804, + "learning_rate": 3.680628529789726e-06, + "loss": 0.8633, + "step": 1721 + }, + { + "epoch": 0.20705825768051464, + "grad_norm": 2.0773852186019965, + "learning_rate": 3.680206122135796e-06, + "loss": 1.1017, + "step": 1722 + }, + { + "epoch": 0.20717850057115372, + "grad_norm": 1.9076250463772795, + "learning_rate": 3.6797834595959323e-06, + "loss": 1.0193, + "step": 1723 + }, + { + "epoch": 0.20729874346179283, + "grad_norm": 4.052148395136947, + "learning_rate": 3.679360542234254e-06, + "loss": 1.0155, + "step": 1724 + }, + { + "epoch": 0.20741898635243192, + "grad_norm": 2.089913340327911, + "learning_rate": 3.678937370114916e-06, + "loss": 0.9644, + "step": 1725 + }, + { + "epoch": 0.207539229243071, + "grad_norm": 3.7838950245754486, + "learning_rate": 3.678513943302114e-06, + "loss": 1.028, + "step": 1726 + }, + { + "epoch": 0.20765947213371008, + "grad_norm": 1.7240944738501838, + "learning_rate": 3.678090261860082e-06, + "loss": 1.0898, + "step": 1727 + }, + { + "epoch": 0.2077797150243492, + "grad_norm": 2.1411623315184047, + "learning_rate": 3.6776663258530906e-06, + "loss": 1.0155, + "step": 1728 + }, + { + "epoch": 0.20789995791498828, + "grad_norm": 2.4498714569002336, + "learning_rate": 3.6772421353454516e-06, + "loss": 0.9511, + "step": 1729 + }, + { + "epoch": 0.20802020080562736, + "grad_norm": 2.432044943168366, + "learning_rate": 3.6768176904015153e-06, + "loss": 1.1203, + "step": 1730 + }, + { + "epoch": 0.20814044369626647, + "grad_norm": 2.8610619743432184, + "learning_rate": 3.6763929910856674e-06, + "loss": 0.8408, + "step": 1731 + }, + { + "epoch": 0.20826068658690555, + "grad_norm": 2.4589265872627943, + "learning_rate": 3.6759680374623365e-06, + "loss": 1.0147, + "step": 1732 + }, + { + "epoch": 0.20838092947754464, + "grad_norm": 2.281132838754827, + "learning_rate": 3.675542829595986e-06, + "loss": 0.9878, + "step": 1733 + }, + { + "epoch": 0.20850117236818372, + "grad_norm": 1.8353515339212312, + "learning_rate": 3.6751173675511213e-06, + "loss": 1.0314, + "step": 1734 + }, + { + "epoch": 0.20862141525882283, + "grad_norm": 2.3196289905321157, + "learning_rate": 3.674691651392283e-06, + "loss": 1.1138, + "step": 1735 + }, + { + "epoch": 0.2087416581494619, + "grad_norm": 2.3300640045019545, + "learning_rate": 3.674265681184053e-06, + "loss": 0.9917, + "step": 1736 + }, + { + "epoch": 0.208861901040101, + "grad_norm": 2.2698823063122187, + "learning_rate": 3.6738394569910504e-06, + "loss": 1.1051, + "step": 1737 + }, + { + "epoch": 0.2089821439307401, + "grad_norm": 2.758882309679652, + "learning_rate": 3.6734129788779333e-06, + "loss": 1.0652, + "step": 1738 + }, + { + "epoch": 0.2091023868213792, + "grad_norm": 1.7996751942572384, + "learning_rate": 3.6729862469093976e-06, + "loss": 1.14, + "step": 1739 + }, + { + "epoch": 0.20922262971201827, + "grad_norm": 2.6751124527726398, + "learning_rate": 3.6725592611501782e-06, + "loss": 1.0693, + "step": 1740 + }, + { + "epoch": 0.20934287260265738, + "grad_norm": 2.7051321526926357, + "learning_rate": 3.6721320216650496e-06, + "loss": 1.004, + "step": 1741 + }, + { + "epoch": 0.20946311549329646, + "grad_norm": 2.021607503043631, + "learning_rate": 3.6717045285188215e-06, + "loss": 1.0753, + "step": 1742 + }, + { + "epoch": 0.20958335838393555, + "grad_norm": 2.001573301430591, + "learning_rate": 3.671276781776346e-06, + "loss": 1.1084, + "step": 1743 + }, + { + "epoch": 0.20970360127457463, + "grad_norm": 2.080041904760941, + "learning_rate": 3.6708487815025128e-06, + "loss": 0.9061, + "step": 1744 + }, + { + "epoch": 0.20982384416521374, + "grad_norm": 3.087604159555748, + "learning_rate": 3.6704205277622463e-06, + "loss": 0.9853, + "step": 1745 + }, + { + "epoch": 0.20994408705585282, + "grad_norm": 2.0355873163920655, + "learning_rate": 3.6699920206205146e-06, + "loss": 1.0459, + "step": 1746 + }, + { + "epoch": 0.2100643299464919, + "grad_norm": 1.8825361753435312, + "learning_rate": 3.669563260142321e-06, + "loss": 1.0577, + "step": 1747 + }, + { + "epoch": 0.21018457283713102, + "grad_norm": 4.682249256432771, + "learning_rate": 3.6691342463927083e-06, + "loss": 1.0855, + "step": 1748 + }, + { + "epoch": 0.2103048157277701, + "grad_norm": 1.8042950224056507, + "learning_rate": 3.668704979436758e-06, + "loss": 1.0576, + "step": 1749 + }, + { + "epoch": 0.21042505861840918, + "grad_norm": 2.1001105415678003, + "learning_rate": 3.668275459339588e-06, + "loss": 1.0279, + "step": 1750 + }, + { + "epoch": 0.21054530150904827, + "grad_norm": 1.9902208981043643, + "learning_rate": 3.667845686166358e-06, + "loss": 1.0409, + "step": 1751 + }, + { + "epoch": 0.21066554439968738, + "grad_norm": 2.010332122237323, + "learning_rate": 3.6674156599822634e-06, + "loss": 1.0986, + "step": 1752 + }, + { + "epoch": 0.21078578729032646, + "grad_norm": 1.9975312612710843, + "learning_rate": 3.666985380852539e-06, + "loss": 1.0555, + "step": 1753 + }, + { + "epoch": 0.21090603018096554, + "grad_norm": 2.2281086154672387, + "learning_rate": 3.6665548488424576e-06, + "loss": 0.9818, + "step": 1754 + }, + { + "epoch": 0.21102627307160465, + "grad_norm": 3.0989065641465454, + "learning_rate": 3.6661240640173307e-06, + "loss": 1.1149, + "step": 1755 + }, + { + "epoch": 0.21114651596224374, + "grad_norm": 0.8998866129021338, + "learning_rate": 3.6656930264425085e-06, + "loss": 0.8169, + "step": 1756 + }, + { + "epoch": 0.21126675885288282, + "grad_norm": 2.007168915312298, + "learning_rate": 3.665261736183378e-06, + "loss": 0.9948, + "step": 1757 + }, + { + "epoch": 0.2113870017435219, + "grad_norm": 3.1305730335372384, + "learning_rate": 3.664830193305366e-06, + "loss": 1.1253, + "step": 1758 + }, + { + "epoch": 0.211507244634161, + "grad_norm": 2.6641926216274747, + "learning_rate": 3.6643983978739373e-06, + "loss": 1.0079, + "step": 1759 + }, + { + "epoch": 0.2116274875248001, + "grad_norm": 2.3184231345223583, + "learning_rate": 3.663966349954596e-06, + "loss": 1.0565, + "step": 1760 + }, + { + "epoch": 0.21174773041543918, + "grad_norm": 0.805051369479276, + "learning_rate": 3.6635340496128816e-06, + "loss": 0.8347, + "step": 1761 + }, + { + "epoch": 0.2118679733060783, + "grad_norm": 1.802129741210797, + "learning_rate": 3.6631014969143747e-06, + "loss": 1.1679, + "step": 1762 + }, + { + "epoch": 0.21198821619671737, + "grad_norm": 1.8304852630222073, + "learning_rate": 3.662668691924693e-06, + "loss": 1.1283, + "step": 1763 + }, + { + "epoch": 0.21210845908735645, + "grad_norm": 2.7892669688268765, + "learning_rate": 3.6622356347094927e-06, + "loss": 0.9529, + "step": 1764 + }, + { + "epoch": 0.21222870197799554, + "grad_norm": 1.9214157548234871, + "learning_rate": 3.6618023253344684e-06, + "loss": 1.026, + "step": 1765 + }, + { + "epoch": 0.21234894486863465, + "grad_norm": 1.7552351982558603, + "learning_rate": 3.6613687638653527e-06, + "loss": 1.0728, + "step": 1766 + }, + { + "epoch": 0.21246918775927373, + "grad_norm": 1.904927154309665, + "learning_rate": 3.660934950367916e-06, + "loss": 1.0177, + "step": 1767 + }, + { + "epoch": 0.21258943064991281, + "grad_norm": 2.040561519646052, + "learning_rate": 3.660500884907968e-06, + "loss": 1.0739, + "step": 1768 + }, + { + "epoch": 0.21270967354055192, + "grad_norm": 0.8452720688830356, + "learning_rate": 3.660066567551356e-06, + "loss": 0.8384, + "step": 1769 + }, + { + "epoch": 0.212829916431191, + "grad_norm": 2.99749396876348, + "learning_rate": 3.6596319983639657e-06, + "loss": 1.0797, + "step": 1770 + }, + { + "epoch": 0.2129501593218301, + "grad_norm": 1.8385645475035446, + "learning_rate": 3.6591971774117214e-06, + "loss": 1.1007, + "step": 1771 + }, + { + "epoch": 0.2130704022124692, + "grad_norm": 1.9573506534357534, + "learning_rate": 3.6587621047605833e-06, + "loss": 1.0391, + "step": 1772 + }, + { + "epoch": 0.21319064510310828, + "grad_norm": 2.601187097539554, + "learning_rate": 3.6583267804765542e-06, + "loss": 1.106, + "step": 1773 + }, + { + "epoch": 0.21331088799374737, + "grad_norm": 2.2766979211288145, + "learning_rate": 3.6578912046256702e-06, + "loss": 1.0969, + "step": 1774 + }, + { + "epoch": 0.21343113088438645, + "grad_norm": 1.9943446429556413, + "learning_rate": 3.6574553772740083e-06, + "loss": 1.0002, + "step": 1775 + }, + { + "epoch": 0.21355137377502556, + "grad_norm": 0.8648803537784453, + "learning_rate": 3.657019298487684e-06, + "loss": 0.8567, + "step": 1776 + }, + { + "epoch": 0.21367161666566464, + "grad_norm": 1.9652719201793327, + "learning_rate": 3.6565829683328495e-06, + "loss": 1.074, + "step": 1777 + }, + { + "epoch": 0.21379185955630373, + "grad_norm": 2.707125487345542, + "learning_rate": 3.6561463868756965e-06, + "loss": 1.1028, + "step": 1778 + }, + { + "epoch": 0.21391210244694284, + "grad_norm": 1.5419103154969054, + "learning_rate": 3.655709554182452e-06, + "loss": 1.0206, + "step": 1779 + }, + { + "epoch": 0.21403234533758192, + "grad_norm": 1.9078814294908297, + "learning_rate": 3.6552724703193855e-06, + "loss": 1.0857, + "step": 1780 + }, + { + "epoch": 0.214152588228221, + "grad_norm": 0.8090209569538903, + "learning_rate": 3.654835135352801e-06, + "loss": 0.797, + "step": 1781 + }, + { + "epoch": 0.21427283111886009, + "grad_norm": 2.0497415286616185, + "learning_rate": 3.654397549349043e-06, + "loss": 1.1156, + "step": 1782 + }, + { + "epoch": 0.2143930740094992, + "grad_norm": 2.5867332317531044, + "learning_rate": 3.653959712374491e-06, + "loss": 0.9908, + "step": 1783 + }, + { + "epoch": 0.21451331690013828, + "grad_norm": 1.917106170297182, + "learning_rate": 3.6535216244955663e-06, + "loss": 1.0667, + "step": 1784 + }, + { + "epoch": 0.21463355979077736, + "grad_norm": 1.8870595936100734, + "learning_rate": 3.653083285778726e-06, + "loss": 0.9508, + "step": 1785 + }, + { + "epoch": 0.21475380268141647, + "grad_norm": 2.058511750471941, + "learning_rate": 3.6526446962904653e-06, + "loss": 1.0488, + "step": 1786 + }, + { + "epoch": 0.21487404557205556, + "grad_norm": 16.523217714259307, + "learning_rate": 3.652205856097318e-06, + "loss": 0.9783, + "step": 1787 + }, + { + "epoch": 0.21499428846269464, + "grad_norm": 1.992939167707856, + "learning_rate": 3.651766765265856e-06, + "loss": 1.031, + "step": 1788 + }, + { + "epoch": 0.21511453135333372, + "grad_norm": 2.388469447294476, + "learning_rate": 3.65132742386269e-06, + "loss": 1.0497, + "step": 1789 + }, + { + "epoch": 0.21523477424397283, + "grad_norm": 2.1624585759599926, + "learning_rate": 3.6508878319544656e-06, + "loss": 1.0868, + "step": 1790 + }, + { + "epoch": 0.21535501713461191, + "grad_norm": 2.7512816130344038, + "learning_rate": 3.65044798960787e-06, + "loss": 1.0554, + "step": 1791 + }, + { + "epoch": 0.215475260025251, + "grad_norm": 1.9224510880140162, + "learning_rate": 3.650007896889627e-06, + "loss": 1.0235, + "step": 1792 + }, + { + "epoch": 0.2155955029158901, + "grad_norm": 2.2934593702584145, + "learning_rate": 3.6495675538664974e-06, + "loss": 1.0458, + "step": 1793 + }, + { + "epoch": 0.2157157458065292, + "grad_norm": 1.7520007549092655, + "learning_rate": 3.649126960605282e-06, + "loss": 1.0632, + "step": 1794 + }, + { + "epoch": 0.21583598869716827, + "grad_norm": 2.6618653127472074, + "learning_rate": 3.6486861171728174e-06, + "loss": 1.0756, + "step": 1795 + }, + { + "epoch": 0.21595623158780738, + "grad_norm": 3.9991969256095263, + "learning_rate": 3.6482450236359803e-06, + "loss": 1.0217, + "step": 1796 + }, + { + "epoch": 0.21607647447844647, + "grad_norm": 2.6773007276165655, + "learning_rate": 3.647803680061683e-06, + "loss": 1.0171, + "step": 1797 + }, + { + "epoch": 0.21619671736908555, + "grad_norm": 2.9772234005253524, + "learning_rate": 3.6473620865168776e-06, + "loss": 0.9921, + "step": 1798 + }, + { + "epoch": 0.21631696025972463, + "grad_norm": 2.053076166649263, + "learning_rate": 3.646920243068554e-06, + "loss": 1.0564, + "step": 1799 + }, + { + "epoch": 0.21643720315036374, + "grad_norm": 3.8881942325395236, + "learning_rate": 3.6464781497837384e-06, + "loss": 0.9848, + "step": 1800 + }, + { + "epoch": 0.21655744604100283, + "grad_norm": 1.7069461897339402, + "learning_rate": 3.6460358067294965e-06, + "loss": 0.9649, + "step": 1801 + }, + { + "epoch": 0.2166776889316419, + "grad_norm": 2.1643678745788084, + "learning_rate": 3.645593213972932e-06, + "loss": 1.0179, + "step": 1802 + }, + { + "epoch": 0.21679793182228102, + "grad_norm": 2.615908364245467, + "learning_rate": 3.6451503715811852e-06, + "loss": 1.0348, + "step": 1803 + }, + { + "epoch": 0.2169181747129201, + "grad_norm": 1.9165498103578957, + "learning_rate": 3.6447072796214345e-06, + "loss": 1.0358, + "step": 1804 + }, + { + "epoch": 0.21703841760355919, + "grad_norm": 0.9675567723160081, + "learning_rate": 3.644263938160898e-06, + "loss": 0.8701, + "step": 1805 + }, + { + "epoch": 0.21715866049419827, + "grad_norm": 1.8322293758014925, + "learning_rate": 3.6438203472668293e-06, + "loss": 0.9533, + "step": 1806 + }, + { + "epoch": 0.21727890338483738, + "grad_norm": 21.997403078261307, + "learning_rate": 3.6433765070065206e-06, + "loss": 1.0586, + "step": 1807 + }, + { + "epoch": 0.21739914627547646, + "grad_norm": 2.289827498889471, + "learning_rate": 3.6429324174473025e-06, + "loss": 1.1149, + "step": 1808 + }, + { + "epoch": 0.21751938916611555, + "grad_norm": 2.742981537626398, + "learning_rate": 3.6424880786565425e-06, + "loss": 1.0883, + "step": 1809 + }, + { + "epoch": 0.21763963205675466, + "grad_norm": 3.503460535727901, + "learning_rate": 3.6420434907016482e-06, + "loss": 1.0393, + "step": 1810 + }, + { + "epoch": 0.21775987494739374, + "grad_norm": 1.9202607361262825, + "learning_rate": 3.6415986536500606e-06, + "loss": 1.0506, + "step": 1811 + }, + { + "epoch": 0.21788011783803282, + "grad_norm": 1.8534433332681715, + "learning_rate": 3.641153567569263e-06, + "loss": 1.0463, + "step": 1812 + }, + { + "epoch": 0.2180003607286719, + "grad_norm": 2.115500365605026, + "learning_rate": 3.640708232526774e-06, + "loss": 1.1959, + "step": 1813 + }, + { + "epoch": 0.21812060361931102, + "grad_norm": 8.523705580540101, + "learning_rate": 3.6402626485901504e-06, + "loss": 1.0217, + "step": 1814 + }, + { + "epoch": 0.2182408465099501, + "grad_norm": 2.682885044474555, + "learning_rate": 3.639816815826988e-06, + "loss": 1.0248, + "step": 1815 + }, + { + "epoch": 0.21836108940058918, + "grad_norm": 2.2246623179471627, + "learning_rate": 3.6393707343049176e-06, + "loss": 1.0187, + "step": 1816 + }, + { + "epoch": 0.2184813322912283, + "grad_norm": 2.776880153159255, + "learning_rate": 3.6389244040916104e-06, + "loss": 0.9737, + "step": 1817 + }, + { + "epoch": 0.21860157518186737, + "grad_norm": 2.123704403123397, + "learning_rate": 3.6384778252547747e-06, + "loss": 1.0353, + "step": 1818 + }, + { + "epoch": 0.21872181807250646, + "grad_norm": 3.042246427979207, + "learning_rate": 3.638030997862155e-06, + "loss": 1.0164, + "step": 1819 + }, + { + "epoch": 0.21884206096314554, + "grad_norm": 0.7769713923712336, + "learning_rate": 3.6375839219815356e-06, + "loss": 0.8324, + "step": 1820 + }, + { + "epoch": 0.21896230385378465, + "grad_norm": 2.1733006251200537, + "learning_rate": 3.6371365976807375e-06, + "loss": 1.0699, + "step": 1821 + }, + { + "epoch": 0.21908254674442373, + "grad_norm": 2.119584419435683, + "learning_rate": 3.6366890250276185e-06, + "loss": 1.0746, + "step": 1822 + }, + { + "epoch": 0.21920278963506282, + "grad_norm": 2.323199290600713, + "learning_rate": 3.6362412040900764e-06, + "loss": 1.1406, + "step": 1823 + }, + { + "epoch": 0.21932303252570193, + "grad_norm": 2.073068878361829, + "learning_rate": 3.635793134936044e-06, + "loss": 1.0444, + "step": 1824 + }, + { + "epoch": 0.219443275416341, + "grad_norm": 1.9602796203298067, + "learning_rate": 3.635344817633494e-06, + "loss": 0.9693, + "step": 1825 + }, + { + "epoch": 0.2195635183069801, + "grad_norm": 2.4324547810286767, + "learning_rate": 3.634896252250436e-06, + "loss": 0.9955, + "step": 1826 + }, + { + "epoch": 0.2196837611976192, + "grad_norm": 1.7643913077486895, + "learning_rate": 3.6344474388549157e-06, + "loss": 1.0635, + "step": 1827 + }, + { + "epoch": 0.2198040040882583, + "grad_norm": 2.368844135598061, + "learning_rate": 3.6339983775150183e-06, + "loss": 1.0422, + "step": 1828 + }, + { + "epoch": 0.21992424697889737, + "grad_norm": 3.6647183414910534, + "learning_rate": 3.6335490682988664e-06, + "loss": 1.0829, + "step": 1829 + }, + { + "epoch": 0.22004448986953645, + "grad_norm": 2.4281958021055545, + "learning_rate": 3.63309951127462e-06, + "loss": 1.0655, + "step": 1830 + }, + { + "epoch": 0.22016473276017556, + "grad_norm": 2.8337701759915808, + "learning_rate": 3.6326497065104757e-06, + "loss": 0.995, + "step": 1831 + }, + { + "epoch": 0.22028497565081465, + "grad_norm": 2.5282713707422686, + "learning_rate": 3.6321996540746697e-06, + "loss": 1.0148, + "step": 1832 + }, + { + "epoch": 0.22040521854145373, + "grad_norm": 2.06225492003889, + "learning_rate": 3.6317493540354733e-06, + "loss": 1.0506, + "step": 1833 + }, + { + "epoch": 0.22052546143209284, + "grad_norm": 2.3489811819336937, + "learning_rate": 3.6312988064611976e-06, + "loss": 1.0112, + "step": 1834 + }, + { + "epoch": 0.22064570432273192, + "grad_norm": 1.729576483577924, + "learning_rate": 3.6308480114201896e-06, + "loss": 1.0524, + "step": 1835 + }, + { + "epoch": 0.220765947213371, + "grad_norm": 1.9101015796911873, + "learning_rate": 3.630396968980835e-06, + "loss": 1.0046, + "step": 1836 + }, + { + "epoch": 0.2208861901040101, + "grad_norm": 2.729013297937968, + "learning_rate": 3.6299456792115575e-06, + "loss": 1.0718, + "step": 1837 + }, + { + "epoch": 0.2210064329946492, + "grad_norm": 1.8327491508651554, + "learning_rate": 3.629494142180815e-06, + "loss": 1.0496, + "step": 1838 + }, + { + "epoch": 0.22112667588528828, + "grad_norm": 2.2191533541402437, + "learning_rate": 3.6290423579571075e-06, + "loss": 1.0886, + "step": 1839 + }, + { + "epoch": 0.22124691877592736, + "grad_norm": 1.768702298117593, + "learning_rate": 3.6285903266089694e-06, + "loss": 1.0411, + "step": 1840 + }, + { + "epoch": 0.22136716166656648, + "grad_norm": 3.271295324528307, + "learning_rate": 3.628138048204974e-06, + "loss": 1.0088, + "step": 1841 + }, + { + "epoch": 0.22148740455720556, + "grad_norm": 1.9923429540680881, + "learning_rate": 3.6276855228137304e-06, + "loss": 1.0027, + "step": 1842 + }, + { + "epoch": 0.22160764744784464, + "grad_norm": 3.095235448703403, + "learning_rate": 3.6272327505038874e-06, + "loss": 1.0607, + "step": 1843 + }, + { + "epoch": 0.22172789033848372, + "grad_norm": 2.007427965441846, + "learning_rate": 3.626779731344131e-06, + "loss": 1.0223, + "step": 1844 + }, + { + "epoch": 0.22184813322912283, + "grad_norm": 2.2197229105914746, + "learning_rate": 3.6263264654031814e-06, + "loss": 1.0906, + "step": 1845 + }, + { + "epoch": 0.22196837611976192, + "grad_norm": 0.7069992564007249, + "learning_rate": 3.6258729527498008e-06, + "loss": 0.8299, + "step": 1846 + }, + { + "epoch": 0.222088619010401, + "grad_norm": 2.443096290043001, + "learning_rate": 3.6254191934527854e-06, + "loss": 0.8814, + "step": 1847 + }, + { + "epoch": 0.2222088619010401, + "grad_norm": 3.3027901038504965, + "learning_rate": 3.6249651875809715e-06, + "loss": 0.8863, + "step": 1848 + }, + { + "epoch": 0.2223291047916792, + "grad_norm": 2.1516155250237423, + "learning_rate": 3.62451093520323e-06, + "loss": 1.1314, + "step": 1849 + }, + { + "epoch": 0.22244934768231828, + "grad_norm": 2.2743768488327945, + "learning_rate": 3.6240564363884714e-06, + "loss": 1.1391, + "step": 1850 + }, + { + "epoch": 0.2225695905729574, + "grad_norm": 2.2490353105689636, + "learning_rate": 3.623601691205643e-06, + "loss": 0.9427, + "step": 1851 + }, + { + "epoch": 0.22268983346359647, + "grad_norm": 2.054930349872805, + "learning_rate": 3.623146699723729e-06, + "loss": 1.0535, + "step": 1852 + }, + { + "epoch": 0.22281007635423555, + "grad_norm": 5.381525381310349, + "learning_rate": 3.6226914620117507e-06, + "loss": 1.0149, + "step": 1853 + }, + { + "epoch": 0.22293031924487464, + "grad_norm": 4.469546160433856, + "learning_rate": 3.622235978138768e-06, + "loss": 1.0477, + "step": 1854 + }, + { + "epoch": 0.22305056213551375, + "grad_norm": 1.9359600346002648, + "learning_rate": 3.621780248173877e-06, + "loss": 1.0524, + "step": 1855 + }, + { + "epoch": 0.22317080502615283, + "grad_norm": 0.8303598441975099, + "learning_rate": 3.6213242721862125e-06, + "loss": 0.8489, + "step": 1856 + }, + { + "epoch": 0.2232910479167919, + "grad_norm": 2.5010300421666405, + "learning_rate": 3.620868050244945e-06, + "loss": 0.9881, + "step": 1857 + }, + { + "epoch": 0.22341129080743102, + "grad_norm": 2.2942000402352623, + "learning_rate": 3.6204115824192817e-06, + "loss": 1.0146, + "step": 1858 + }, + { + "epoch": 0.2235315336980701, + "grad_norm": 5.256764368388207, + "learning_rate": 3.619954868778471e-06, + "loss": 1.0076, + "step": 1859 + }, + { + "epoch": 0.2236517765887092, + "grad_norm": 2.0354496896252092, + "learning_rate": 3.6194979093917944e-06, + "loss": 1.0696, + "step": 1860 + }, + { + "epoch": 0.22377201947934827, + "grad_norm": 2.758572136106384, + "learning_rate": 3.6190407043285724e-06, + "loss": 1.1084, + "step": 1861 + }, + { + "epoch": 0.22389226236998738, + "grad_norm": 1.847574695940621, + "learning_rate": 3.618583253658163e-06, + "loss": 0.9965, + "step": 1862 + }, + { + "epoch": 0.22401250526062647, + "grad_norm": 2.0472543379485773, + "learning_rate": 3.618125557449961e-06, + "loss": 1.1063, + "step": 1863 + }, + { + "epoch": 0.22413274815126555, + "grad_norm": 1.9714652819206844, + "learning_rate": 3.6176676157733983e-06, + "loss": 1.0671, + "step": 1864 + }, + { + "epoch": 0.22425299104190466, + "grad_norm": 2.0552010130200054, + "learning_rate": 3.6172094286979443e-06, + "loss": 1.0016, + "step": 1865 + }, + { + "epoch": 0.22437323393254374, + "grad_norm": 1.869450239569625, + "learning_rate": 3.6167509962931064e-06, + "loss": 1.0518, + "step": 1866 + }, + { + "epoch": 0.22449347682318282, + "grad_norm": 2.717341354076888, + "learning_rate": 3.6162923186284276e-06, + "loss": 1.0052, + "step": 1867 + }, + { + "epoch": 0.2246137197138219, + "grad_norm": 2.392688135225979, + "learning_rate": 3.6158333957734888e-06, + "loss": 1.0969, + "step": 1868 + }, + { + "epoch": 0.22473396260446102, + "grad_norm": 2.2038597883636695, + "learning_rate": 3.6153742277979088e-06, + "loss": 1.0653, + "step": 1869 + }, + { + "epoch": 0.2248542054951001, + "grad_norm": 2.2679583931324085, + "learning_rate": 3.6149148147713434e-06, + "loss": 1.015, + "step": 1870 + }, + { + "epoch": 0.22497444838573918, + "grad_norm": 2.473713195434467, + "learning_rate": 3.614455156763484e-06, + "loss": 1.1016, + "step": 1871 + }, + { + "epoch": 0.2250946912763783, + "grad_norm": 4.5347529058633755, + "learning_rate": 3.613995253844061e-06, + "loss": 0.9549, + "step": 1872 + }, + { + "epoch": 0.22521493416701738, + "grad_norm": 11.66654648491856, + "learning_rate": 3.6135351060828414e-06, + "loss": 1.0437, + "step": 1873 + }, + { + "epoch": 0.22533517705765646, + "grad_norm": 4.707244994822787, + "learning_rate": 3.6130747135496285e-06, + "loss": 0.9304, + "step": 1874 + }, + { + "epoch": 0.22545541994829554, + "grad_norm": 1.974862555055071, + "learning_rate": 3.6126140763142646e-06, + "loss": 0.897, + "step": 1875 + }, + { + "epoch": 0.22557566283893465, + "grad_norm": 5.076682487424735, + "learning_rate": 3.6121531944466275e-06, + "loss": 1.1025, + "step": 1876 + }, + { + "epoch": 0.22569590572957374, + "grad_norm": 2.6889179835921646, + "learning_rate": 3.611692068016633e-06, + "loss": 1.0163, + "step": 1877 + }, + { + "epoch": 0.22581614862021282, + "grad_norm": 2.5360677576369453, + "learning_rate": 3.611230697094233e-06, + "loss": 0.9859, + "step": 1878 + }, + { + "epoch": 0.22593639151085193, + "grad_norm": 2.2506323031895183, + "learning_rate": 3.6107690817494173e-06, + "loss": 1.1088, + "step": 1879 + }, + { + "epoch": 0.226056634401491, + "grad_norm": 4.21385032676957, + "learning_rate": 3.6103072220522117e-06, + "loss": 0.9465, + "step": 1880 + }, + { + "epoch": 0.2261768772921301, + "grad_norm": 2.6598238072087192, + "learning_rate": 3.609845118072682e-06, + "loss": 1.1552, + "step": 1881 + }, + { + "epoch": 0.2262971201827692, + "grad_norm": 2.326104330496326, + "learning_rate": 3.6093827698809276e-06, + "loss": 1.0341, + "step": 1882 + }, + { + "epoch": 0.2264173630734083, + "grad_norm": 2.4952713114397183, + "learning_rate": 3.6089201775470864e-06, + "loss": 1.0849, + "step": 1883 + }, + { + "epoch": 0.22653760596404737, + "grad_norm": 1.5552517427729948, + "learning_rate": 3.6084573411413334e-06, + "loss": 1.0131, + "step": 1884 + }, + { + "epoch": 0.22665784885468646, + "grad_norm": 2.710818593168558, + "learning_rate": 3.607994260733881e-06, + "loss": 1.0487, + "step": 1885 + }, + { + "epoch": 0.22677809174532557, + "grad_norm": 1.8934991601177051, + "learning_rate": 3.6075309363949776e-06, + "loss": 0.9796, + "step": 1886 + }, + { + "epoch": 0.22689833463596465, + "grad_norm": 2.1707146927623326, + "learning_rate": 3.6070673681949094e-06, + "loss": 1.048, + "step": 1887 + }, + { + "epoch": 0.22701857752660373, + "grad_norm": 1.9430642368733806, + "learning_rate": 3.606603556203999e-06, + "loss": 1.0534, + "step": 1888 + }, + { + "epoch": 0.22713882041724284, + "grad_norm": 2.0495890321734413, + "learning_rate": 3.6061395004926066e-06, + "loss": 1.0743, + "step": 1889 + }, + { + "epoch": 0.22725906330788193, + "grad_norm": 2.400805965833869, + "learning_rate": 3.605675201131129e-06, + "loss": 1.0854, + "step": 1890 + }, + { + "epoch": 0.227379306198521, + "grad_norm": 2.3315690954851593, + "learning_rate": 3.60521065819e-06, + "loss": 1.0343, + "step": 1891 + }, + { + "epoch": 0.2274995490891601, + "grad_norm": 2.438650397725984, + "learning_rate": 3.60474587173969e-06, + "loss": 1.1162, + "step": 1892 + }, + { + "epoch": 0.2276197919797992, + "grad_norm": 2.1122316585055523, + "learning_rate": 3.6042808418507084e-06, + "loss": 1.0787, + "step": 1893 + }, + { + "epoch": 0.22774003487043828, + "grad_norm": 2.394824758342235, + "learning_rate": 3.6038155685935976e-06, + "loss": 1.0117, + "step": 1894 + }, + { + "epoch": 0.22786027776107737, + "grad_norm": 2.320353806253229, + "learning_rate": 3.6033500520389404e-06, + "loss": 0.9424, + "step": 1895 + }, + { + "epoch": 0.22798052065171648, + "grad_norm": 0.7992048797447947, + "learning_rate": 3.6028842922573553e-06, + "loss": 0.8862, + "step": 1896 + }, + { + "epoch": 0.22810076354235556, + "grad_norm": 0.8761762818741402, + "learning_rate": 3.602418289319497e-06, + "loss": 0.8673, + "step": 1897 + }, + { + "epoch": 0.22822100643299464, + "grad_norm": 1.784110399392669, + "learning_rate": 3.601952043296059e-06, + "loss": 0.9759, + "step": 1898 + }, + { + "epoch": 0.22834124932363373, + "grad_norm": 2.5586562753273867, + "learning_rate": 3.6014855542577696e-06, + "loss": 1.0446, + "step": 1899 + }, + { + "epoch": 0.22846149221427284, + "grad_norm": 1.8853330084293702, + "learning_rate": 3.6010188222753943e-06, + "loss": 1.0801, + "step": 1900 + }, + { + "epoch": 0.22858173510491192, + "grad_norm": 0.9121839341724168, + "learning_rate": 3.6005518474197372e-06, + "loss": 0.8797, + "step": 1901 + }, + { + "epoch": 0.228701977995551, + "grad_norm": 1.8277503387738878, + "learning_rate": 3.6000846297616373e-06, + "loss": 1.0226, + "step": 1902 + }, + { + "epoch": 0.22882222088619011, + "grad_norm": 2.4912857768183367, + "learning_rate": 3.5996171693719717e-06, + "loss": 0.9673, + "step": 1903 + }, + { + "epoch": 0.2289424637768292, + "grad_norm": 0.8437396508041333, + "learning_rate": 3.5991494663216528e-06, + "loss": 0.8862, + "step": 1904 + }, + { + "epoch": 0.22906270666746828, + "grad_norm": 3.3656628336502235, + "learning_rate": 3.5986815206816314e-06, + "loss": 1.1113, + "step": 1905 + }, + { + "epoch": 0.2291829495581074, + "grad_norm": 2.257318356832509, + "learning_rate": 3.598213332522895e-06, + "loss": 0.9872, + "step": 1906 + }, + { + "epoch": 0.22930319244874647, + "grad_norm": 2.0982370833343973, + "learning_rate": 3.597744901916466e-06, + "loss": 1.0133, + "step": 1907 + }, + { + "epoch": 0.22942343533938556, + "grad_norm": 2.7203907893449664, + "learning_rate": 3.5972762289334058e-06, + "loss": 1.0082, + "step": 1908 + }, + { + "epoch": 0.22954367823002464, + "grad_norm": 2.0200748265448207, + "learning_rate": 3.5968073136448116e-06, + "loss": 1.0896, + "step": 1909 + }, + { + "epoch": 0.22966392112066375, + "grad_norm": 4.3184183393616875, + "learning_rate": 3.596338156121818e-06, + "loss": 1.1519, + "step": 1910 + }, + { + "epoch": 0.22978416401130283, + "grad_norm": 0.7470637701408792, + "learning_rate": 3.595868756435595e-06, + "loss": 0.8313, + "step": 1911 + }, + { + "epoch": 0.22990440690194192, + "grad_norm": 3.246686126238126, + "learning_rate": 3.5953991146573504e-06, + "loss": 1.0456, + "step": 1912 + }, + { + "epoch": 0.23002464979258103, + "grad_norm": 2.6403914128205934, + "learning_rate": 3.5949292308583294e-06, + "loss": 1.071, + "step": 1913 + }, + { + "epoch": 0.2301448926832201, + "grad_norm": 2.2492571770160374, + "learning_rate": 3.594459105109811e-06, + "loss": 1.051, + "step": 1914 + }, + { + "epoch": 0.2302651355738592, + "grad_norm": 1.8799743788576302, + "learning_rate": 3.593988737483115e-06, + "loss": 1.0513, + "step": 1915 + }, + { + "epoch": 0.23038537846449827, + "grad_norm": 2.402379676116024, + "learning_rate": 3.5935181280495947e-06, + "loss": 1.0198, + "step": 1916 + }, + { + "epoch": 0.23050562135513739, + "grad_norm": 0.7994723904764662, + "learning_rate": 3.5930472768806412e-06, + "loss": 0.7808, + "step": 1917 + }, + { + "epoch": 0.23062586424577647, + "grad_norm": 2.7024773499900925, + "learning_rate": 3.5925761840476826e-06, + "loss": 1.0115, + "step": 1918 + }, + { + "epoch": 0.23074610713641555, + "grad_norm": 5.319834593876126, + "learning_rate": 3.592104849622183e-06, + "loss": 1.0543, + "step": 1919 + }, + { + "epoch": 0.23086635002705466, + "grad_norm": 1.4869703391682163, + "learning_rate": 3.591633273675644e-06, + "loss": 0.9726, + "step": 1920 + }, + { + "epoch": 0.23098659291769374, + "grad_norm": 0.9524013085687548, + "learning_rate": 3.591161456279602e-06, + "loss": 0.8196, + "step": 1921 + }, + { + "epoch": 0.23110683580833283, + "grad_norm": 1.5131544447014975, + "learning_rate": 3.590689397505633e-06, + "loss": 1.0398, + "step": 1922 + }, + { + "epoch": 0.2312270786989719, + "grad_norm": 1.8033369551297318, + "learning_rate": 3.590217097425347e-06, + "loss": 1.104, + "step": 1923 + }, + { + "epoch": 0.23134732158961102, + "grad_norm": 4.085215284074553, + "learning_rate": 3.589744556110391e-06, + "loss": 0.9479, + "step": 1924 + }, + { + "epoch": 0.2314675644802501, + "grad_norm": 1.7716406234927597, + "learning_rate": 3.58927177363245e-06, + "loss": 1.0859, + "step": 1925 + }, + { + "epoch": 0.2315878073708892, + "grad_norm": 4.239062877954508, + "learning_rate": 3.5887987500632447e-06, + "loss": 0.9559, + "step": 1926 + }, + { + "epoch": 0.2317080502615283, + "grad_norm": 1.9012061130692663, + "learning_rate": 3.5883254854745325e-06, + "loss": 1.077, + "step": 1927 + }, + { + "epoch": 0.23182829315216738, + "grad_norm": 2.270542989478846, + "learning_rate": 3.587851979938107e-06, + "loss": 0.9902, + "step": 1928 + }, + { + "epoch": 0.23194853604280646, + "grad_norm": 2.0684248469925732, + "learning_rate": 3.5873782335257985e-06, + "loss": 1.0136, + "step": 1929 + }, + { + "epoch": 0.23206877893344555, + "grad_norm": 2.4398411975838696, + "learning_rate": 3.5869042463094744e-06, + "loss": 1.0194, + "step": 1930 + }, + { + "epoch": 0.23218902182408466, + "grad_norm": 2.1678118124855317, + "learning_rate": 3.586430018361038e-06, + "loss": 1.0093, + "step": 1931 + }, + { + "epoch": 0.23230926471472374, + "grad_norm": 3.5691819018147823, + "learning_rate": 3.5859555497524283e-06, + "loss": 1.0024, + "step": 1932 + }, + { + "epoch": 0.23242950760536282, + "grad_norm": 2.446888863437788, + "learning_rate": 3.5854808405556237e-06, + "loss": 1.1611, + "step": 1933 + }, + { + "epoch": 0.23254975049600193, + "grad_norm": 2.3175956670115547, + "learning_rate": 3.5850058908426355e-06, + "loss": 0.9906, + "step": 1934 + }, + { + "epoch": 0.23266999338664102, + "grad_norm": 1.869511327386306, + "learning_rate": 3.584530700685514e-06, + "loss": 1.093, + "step": 1935 + }, + { + "epoch": 0.2327902362772801, + "grad_norm": 2.4296674819563226, + "learning_rate": 3.5840552701563448e-06, + "loss": 1.1282, + "step": 1936 + }, + { + "epoch": 0.2329104791679192, + "grad_norm": 2.896878134058108, + "learning_rate": 3.5835795993272513e-06, + "loss": 1.0547, + "step": 1937 + }, + { + "epoch": 0.2330307220585583, + "grad_norm": 2.114156206962424, + "learning_rate": 3.583103688270391e-06, + "loss": 0.9492, + "step": 1938 + }, + { + "epoch": 0.23315096494919738, + "grad_norm": 2.4080720418754717, + "learning_rate": 3.58262753705796e-06, + "loss": 1.1347, + "step": 1939 + }, + { + "epoch": 0.23327120783983646, + "grad_norm": 0.7623799120032841, + "learning_rate": 3.5821511457621902e-06, + "loss": 0.7935, + "step": 1940 + }, + { + "epoch": 0.23339145073047557, + "grad_norm": 3.512917251511455, + "learning_rate": 3.5816745144553497e-06, + "loss": 1.0502, + "step": 1941 + }, + { + "epoch": 0.23351169362111465, + "grad_norm": 3.1154230999358288, + "learning_rate": 3.5811976432097424e-06, + "loss": 0.9921, + "step": 1942 + }, + { + "epoch": 0.23363193651175373, + "grad_norm": 4.278459070842448, + "learning_rate": 3.58072053209771e-06, + "loss": 1.0856, + "step": 1943 + }, + { + "epoch": 0.23375217940239285, + "grad_norm": 3.701616970062435, + "learning_rate": 3.5802431811916296e-06, + "loss": 1.0287, + "step": 1944 + }, + { + "epoch": 0.23387242229303193, + "grad_norm": 1.9141724846828572, + "learning_rate": 3.579765590563916e-06, + "loss": 1.0422, + "step": 1945 + }, + { + "epoch": 0.233992665183671, + "grad_norm": 3.662587208409392, + "learning_rate": 3.579287760287017e-06, + "loss": 1.0551, + "step": 1946 + }, + { + "epoch": 0.2341129080743101, + "grad_norm": 1.7536021353391886, + "learning_rate": 3.578809690433421e-06, + "loss": 0.9664, + "step": 1947 + }, + { + "epoch": 0.2342331509649492, + "grad_norm": 2.2748052408881443, + "learning_rate": 3.578331381075651e-06, + "loss": 1.0521, + "step": 1948 + }, + { + "epoch": 0.2343533938555883, + "grad_norm": 2.288091251694373, + "learning_rate": 3.5778528322862646e-06, + "loss": 0.9313, + "step": 1949 + }, + { + "epoch": 0.23447363674622737, + "grad_norm": 2.0141301489242, + "learning_rate": 3.5773740441378585e-06, + "loss": 1.1057, + "step": 1950 + }, + { + "epoch": 0.23459387963686648, + "grad_norm": 1.8339045530585838, + "learning_rate": 3.5768950167030633e-06, + "loss": 0.9772, + "step": 1951 + }, + { + "epoch": 0.23471412252750556, + "grad_norm": 1.98816948896972, + "learning_rate": 3.576415750054548e-06, + "loss": 1.0207, + "step": 1952 + }, + { + "epoch": 0.23483436541814465, + "grad_norm": 2.6806806942505457, + "learning_rate": 3.5759362442650172e-06, + "loss": 1.0906, + "step": 1953 + }, + { + "epoch": 0.23495460830878373, + "grad_norm": 2.9461885560935572, + "learning_rate": 3.5754564994072113e-06, + "loss": 1.0893, + "step": 1954 + }, + { + "epoch": 0.23507485119942284, + "grad_norm": 2.3455719033667517, + "learning_rate": 3.5749765155539067e-06, + "loss": 0.8346, + "step": 1955 + }, + { + "epoch": 0.23519509409006192, + "grad_norm": 14.669013153563332, + "learning_rate": 3.574496292777917e-06, + "loss": 1.166, + "step": 1956 + }, + { + "epoch": 0.235315336980701, + "grad_norm": 2.879742401311721, + "learning_rate": 3.574015831152092e-06, + "loss": 0.9541, + "step": 1957 + }, + { + "epoch": 0.23543557987134012, + "grad_norm": 4.032090446926059, + "learning_rate": 3.573535130749316e-06, + "loss": 1.075, + "step": 1958 + }, + { + "epoch": 0.2355558227619792, + "grad_norm": 1.8395201445023726, + "learning_rate": 3.5730541916425127e-06, + "loss": 0.9784, + "step": 1959 + }, + { + "epoch": 0.23567606565261828, + "grad_norm": 2.4531281100696845, + "learning_rate": 3.572573013904639e-06, + "loss": 1.1011, + "step": 1960 + }, + { + "epoch": 0.2357963085432574, + "grad_norm": 2.0345927975937257, + "learning_rate": 3.572091597608689e-06, + "loss": 1.158, + "step": 1961 + }, + { + "epoch": 0.23591655143389648, + "grad_norm": 2.742135941327752, + "learning_rate": 3.571609942827694e-06, + "loss": 0.9745, + "step": 1962 + }, + { + "epoch": 0.23603679432453556, + "grad_norm": 1.8495252644508977, + "learning_rate": 3.57112804963472e-06, + "loss": 1.1233, + "step": 1963 + }, + { + "epoch": 0.23615703721517464, + "grad_norm": 1.7558887042183404, + "learning_rate": 3.57064591810287e-06, + "loss": 1.0031, + "step": 1964 + }, + { + "epoch": 0.23627728010581375, + "grad_norm": 3.1194095197084897, + "learning_rate": 3.570163548305284e-06, + "loss": 1.0419, + "step": 1965 + }, + { + "epoch": 0.23639752299645284, + "grad_norm": 3.4959318496468845, + "learning_rate": 3.569680940315135e-06, + "loss": 0.9424, + "step": 1966 + }, + { + "epoch": 0.23651776588709192, + "grad_norm": 3.574826114934006, + "learning_rate": 3.5691980942056356e-06, + "loss": 1.0601, + "step": 1967 + }, + { + "epoch": 0.23663800877773103, + "grad_norm": 3.802229257509357, + "learning_rate": 3.5687150100500332e-06, + "loss": 1.0377, + "step": 1968 + }, + { + "epoch": 0.2367582516683701, + "grad_norm": 1.937365988742116, + "learning_rate": 3.568231687921611e-06, + "loss": 0.9817, + "step": 1969 + }, + { + "epoch": 0.2368784945590092, + "grad_norm": 1.6029564722425549, + "learning_rate": 3.5677481278936883e-06, + "loss": 1.0427, + "step": 1970 + }, + { + "epoch": 0.23699873744964828, + "grad_norm": 0.8400217266905993, + "learning_rate": 3.5672643300396214e-06, + "loss": 0.8166, + "step": 1971 + }, + { + "epoch": 0.2371189803402874, + "grad_norm": 2.3198282783358763, + "learning_rate": 3.566780294432802e-06, + "loss": 0.9171, + "step": 1972 + }, + { + "epoch": 0.23723922323092647, + "grad_norm": 3.088380257225495, + "learning_rate": 3.566296021146657e-06, + "loss": 0.9858, + "step": 1973 + }, + { + "epoch": 0.23735946612156555, + "grad_norm": 1.60350083078304, + "learning_rate": 3.565811510254652e-06, + "loss": 0.971, + "step": 1974 + }, + { + "epoch": 0.23747970901220466, + "grad_norm": 0.8280983686663477, + "learning_rate": 3.5653267618302845e-06, + "loss": 0.8203, + "step": 1975 + }, + { + "epoch": 0.23759995190284375, + "grad_norm": 1.9334462119177953, + "learning_rate": 3.564841775947093e-06, + "loss": 1.0933, + "step": 1976 + }, + { + "epoch": 0.23772019479348283, + "grad_norm": 2.208362387220126, + "learning_rate": 3.5643565526786475e-06, + "loss": 0.9941, + "step": 1977 + }, + { + "epoch": 0.2378404376841219, + "grad_norm": 1.8925701255353709, + "learning_rate": 3.5638710920985574e-06, + "loss": 1.0105, + "step": 1978 + }, + { + "epoch": 0.23796068057476102, + "grad_norm": 1.976910466259832, + "learning_rate": 3.5633853942804655e-06, + "loss": 1.0638, + "step": 1979 + }, + { + "epoch": 0.2380809234654001, + "grad_norm": 2.173269249706527, + "learning_rate": 3.5628994592980527e-06, + "loss": 0.9986, + "step": 1980 + }, + { + "epoch": 0.2382011663560392, + "grad_norm": 3.6498433902534044, + "learning_rate": 3.562413287225034e-06, + "loss": 0.9407, + "step": 1981 + }, + { + "epoch": 0.2383214092466783, + "grad_norm": 2.686745116206226, + "learning_rate": 3.5619268781351623e-06, + "loss": 1.1305, + "step": 1982 + }, + { + "epoch": 0.23844165213731738, + "grad_norm": 1.9249576563946422, + "learning_rate": 3.5614402321022256e-06, + "loss": 1.0057, + "step": 1983 + }, + { + "epoch": 0.23856189502795647, + "grad_norm": 2.2971103054626365, + "learning_rate": 3.5609533492000463e-06, + "loss": 1.1062, + "step": 1984 + }, + { + "epoch": 0.23868213791859555, + "grad_norm": 3.3214718219619015, + "learning_rate": 3.560466229502485e-06, + "loss": 1.0207, + "step": 1985 + }, + { + "epoch": 0.23880238080923466, + "grad_norm": 2.270039224159713, + "learning_rate": 3.5599788730834384e-06, + "loss": 1.1359, + "step": 1986 + }, + { + "epoch": 0.23892262369987374, + "grad_norm": 2.556539068689369, + "learning_rate": 3.559491280016836e-06, + "loss": 1.0254, + "step": 1987 + }, + { + "epoch": 0.23904286659051283, + "grad_norm": 2.1434783852450425, + "learning_rate": 3.5590034503766465e-06, + "loss": 0.9495, + "step": 1988 + }, + { + "epoch": 0.23916310948115194, + "grad_norm": 2.6157815829388995, + "learning_rate": 3.558515384236874e-06, + "loss": 1.0514, + "step": 1989 + }, + { + "epoch": 0.23928335237179102, + "grad_norm": 2.130851878459831, + "learning_rate": 3.558027081671556e-06, + "loss": 1.0774, + "step": 1990 + }, + { + "epoch": 0.2394035952624301, + "grad_norm": 1.9164540822544245, + "learning_rate": 3.557538542754769e-06, + "loss": 0.924, + "step": 1991 + }, + { + "epoch": 0.2395238381530692, + "grad_norm": 2.2116595595513195, + "learning_rate": 3.557049767560623e-06, + "loss": 0.9107, + "step": 1992 + }, + { + "epoch": 0.2396440810437083, + "grad_norm": 2.4632450970998057, + "learning_rate": 3.5565607561632655e-06, + "loss": 1.1006, + "step": 1993 + }, + { + "epoch": 0.23976432393434738, + "grad_norm": 2.2923382439741253, + "learning_rate": 3.5560715086368787e-06, + "loss": 1.0315, + "step": 1994 + }, + { + "epoch": 0.23988456682498646, + "grad_norm": 2.8670296690529695, + "learning_rate": 3.5555820250556816e-06, + "loss": 1.0644, + "step": 1995 + }, + { + "epoch": 0.24000480971562557, + "grad_norm": 2.7435289020131792, + "learning_rate": 3.5550923054939278e-06, + "loss": 0.9312, + "step": 1996 + }, + { + "epoch": 0.24012505260626466, + "grad_norm": 2.066521855890129, + "learning_rate": 3.5546023500259083e-06, + "loss": 0.9846, + "step": 1997 + }, + { + "epoch": 0.24024529549690374, + "grad_norm": 2.2170057290794256, + "learning_rate": 3.5541121587259477e-06, + "loss": 1.0419, + "step": 1998 + }, + { + "epoch": 0.24036553838754285, + "grad_norm": 0.8542751537992561, + "learning_rate": 3.553621731668408e-06, + "loss": 0.817, + "step": 1999 + }, + { + "epoch": 0.24048578127818193, + "grad_norm": 1.974406880033491, + "learning_rate": 3.553131068927688e-06, + "loss": 1.0734, + "step": 2000 + }, + { + "epoch": 0.24060602416882101, + "grad_norm": 1.9058903839597297, + "learning_rate": 3.552640170578219e-06, + "loss": 1.0393, + "step": 2001 + }, + { + "epoch": 0.2407262670594601, + "grad_norm": 1.9607225885446227, + "learning_rate": 3.5521490366944703e-06, + "loss": 1.0194, + "step": 2002 + }, + { + "epoch": 0.2408465099500992, + "grad_norm": 2.158305639533608, + "learning_rate": 3.5516576673509474e-06, + "loss": 1.0378, + "step": 2003 + }, + { + "epoch": 0.2409667528407383, + "grad_norm": 1.861061357272795, + "learning_rate": 3.5511660626221896e-06, + "loss": 1.1045, + "step": 2004 + }, + { + "epoch": 0.24108699573137737, + "grad_norm": 3.3314224169916624, + "learning_rate": 3.5506742225827744e-06, + "loss": 1.1327, + "step": 2005 + }, + { + "epoch": 0.24120723862201648, + "grad_norm": 2.171843603991262, + "learning_rate": 3.5501821473073116e-06, + "loss": 1.1396, + "step": 2006 + }, + { + "epoch": 0.24132748151265557, + "grad_norm": 2.2428479434823503, + "learning_rate": 3.54968983687045e-06, + "loss": 1.1066, + "step": 2007 + }, + { + "epoch": 0.24144772440329465, + "grad_norm": 3.5690694101499028, + "learning_rate": 3.549197291346872e-06, + "loss": 1.1305, + "step": 2008 + }, + { + "epoch": 0.24156796729393373, + "grad_norm": 2.0854537345749806, + "learning_rate": 3.548704510811297e-06, + "loss": 1.0303, + "step": 2009 + }, + { + "epoch": 0.24168821018457284, + "grad_norm": 2.7128218112729074, + "learning_rate": 3.5482114953384787e-06, + "loss": 0.9805, + "step": 2010 + }, + { + "epoch": 0.24180845307521193, + "grad_norm": 2.541683593026588, + "learning_rate": 3.5477182450032077e-06, + "loss": 1.086, + "step": 2011 + }, + { + "epoch": 0.241928695965851, + "grad_norm": 2.286690690754993, + "learning_rate": 3.5472247598803097e-06, + "loss": 1.076, + "step": 2012 + }, + { + "epoch": 0.24204893885649012, + "grad_norm": 11.343691486806733, + "learning_rate": 3.546731040044645e-06, + "loss": 1.0958, + "step": 2013 + }, + { + "epoch": 0.2421691817471292, + "grad_norm": 2.0802007647971847, + "learning_rate": 3.546237085571112e-06, + "loss": 0.9887, + "step": 2014 + }, + { + "epoch": 0.24228942463776829, + "grad_norm": 2.04770463044357, + "learning_rate": 3.5457428965346425e-06, + "loss": 0.9572, + "step": 2015 + }, + { + "epoch": 0.2424096675284074, + "grad_norm": 1.6800233017349964, + "learning_rate": 3.545248473010205e-06, + "loss": 0.9868, + "step": 2016 + }, + { + "epoch": 0.24252991041904648, + "grad_norm": 2.683718482704395, + "learning_rate": 3.544753815072802e-06, + "loss": 1.1154, + "step": 2017 + }, + { + "epoch": 0.24265015330968556, + "grad_norm": 2.0201292351654185, + "learning_rate": 3.544258922797474e-06, + "loss": 1.123, + "step": 2018 + }, + { + "epoch": 0.24277039620032465, + "grad_norm": 1.5444304291038158, + "learning_rate": 3.543763796259295e-06, + "loss": 1.0167, + "step": 2019 + }, + { + "epoch": 0.24289063909096376, + "grad_norm": 2.436961187818267, + "learning_rate": 3.5432684355333754e-06, + "loss": 1.1457, + "step": 2020 + }, + { + "epoch": 0.24301088198160284, + "grad_norm": 1.9602555992884259, + "learning_rate": 3.5427728406948613e-06, + "loss": 1.01, + "step": 2021 + }, + { + "epoch": 0.24313112487224192, + "grad_norm": 0.7540368080742184, + "learning_rate": 3.542277011818934e-06, + "loss": 0.818, + "step": 2022 + }, + { + "epoch": 0.24325136776288103, + "grad_norm": 6.70073666294773, + "learning_rate": 3.5417809489808104e-06, + "loss": 0.9828, + "step": 2023 + }, + { + "epoch": 0.24337161065352012, + "grad_norm": 1.9993456723827876, + "learning_rate": 3.5412846522557422e-06, + "loss": 0.9666, + "step": 2024 + }, + { + "epoch": 0.2434918535441592, + "grad_norm": 1.9358205592171485, + "learning_rate": 3.540788121719018e-06, + "loss": 0.983, + "step": 2025 + }, + { + "epoch": 0.24361209643479828, + "grad_norm": 2.0201812819538714, + "learning_rate": 3.5402913574459604e-06, + "loss": 1.0613, + "step": 2026 + }, + { + "epoch": 0.2437323393254374, + "grad_norm": 2.1075086460863828, + "learning_rate": 3.5397943595119297e-06, + "loss": 1.0992, + "step": 2027 + }, + { + "epoch": 0.24385258221607647, + "grad_norm": 2.949471933979209, + "learning_rate": 3.5392971279923177e-06, + "loss": 1.0145, + "step": 2028 + }, + { + "epoch": 0.24397282510671556, + "grad_norm": 2.2140017999850214, + "learning_rate": 3.5387996629625557e-06, + "loss": 1.0667, + "step": 2029 + }, + { + "epoch": 0.24409306799735467, + "grad_norm": 0.8044026435423539, + "learning_rate": 3.5383019644981083e-06, + "loss": 0.7882, + "step": 2030 + }, + { + "epoch": 0.24421331088799375, + "grad_norm": 3.8493532108136765, + "learning_rate": 3.5378040326744763e-06, + "loss": 0.9693, + "step": 2031 + }, + { + "epoch": 0.24433355377863283, + "grad_norm": 2.9163961194217802, + "learning_rate": 3.5373058675671946e-06, + "loss": 1.0959, + "step": 2032 + }, + { + "epoch": 0.24445379666927192, + "grad_norm": 2.1546869555964347, + "learning_rate": 3.536807469251836e-06, + "loss": 0.967, + "step": 2033 + }, + { + "epoch": 0.24457403955991103, + "grad_norm": 2.234390258736961, + "learning_rate": 3.5363088378040055e-06, + "loss": 1.062, + "step": 2034 + }, + { + "epoch": 0.2446942824505501, + "grad_norm": 0.7596012908590114, + "learning_rate": 3.5358099732993463e-06, + "loss": 0.8816, + "step": 2035 + }, + { + "epoch": 0.2448145253411892, + "grad_norm": 4.516553314776085, + "learning_rate": 3.535310875813535e-06, + "loss": 1.1304, + "step": 2036 + }, + { + "epoch": 0.2449347682318283, + "grad_norm": 1.7304721255301798, + "learning_rate": 3.5348115454222843e-06, + "loss": 1.0543, + "step": 2037 + }, + { + "epoch": 0.2450550111224674, + "grad_norm": 3.2667408182200948, + "learning_rate": 3.5343119822013425e-06, + "loss": 1.0956, + "step": 2038 + }, + { + "epoch": 0.24517525401310647, + "grad_norm": 2.173121691425077, + "learning_rate": 3.533812186226493e-06, + "loss": 1.0156, + "step": 2039 + }, + { + "epoch": 0.24529549690374555, + "grad_norm": 2.092128410388086, + "learning_rate": 3.5333121575735545e-06, + "loss": 0.9964, + "step": 2040 + }, + { + "epoch": 0.24541573979438466, + "grad_norm": 2.1111467620560966, + "learning_rate": 3.532811896318381e-06, + "loss": 0.9913, + "step": 2041 + }, + { + "epoch": 0.24553598268502375, + "grad_norm": 2.611156746170185, + "learning_rate": 3.5323114025368615e-06, + "loss": 1.0606, + "step": 2042 + }, + { + "epoch": 0.24565622557566283, + "grad_norm": 3.8128364289477634, + "learning_rate": 3.53181067630492e-06, + "loss": 1.0624, + "step": 2043 + }, + { + "epoch": 0.24577646846630194, + "grad_norm": 1.89560931492626, + "learning_rate": 3.5313097176985175e-06, + "loss": 1.0009, + "step": 2044 + }, + { + "epoch": 0.24589671135694102, + "grad_norm": 2.727631542414913, + "learning_rate": 3.5308085267936482e-06, + "loss": 1.0527, + "step": 2045 + }, + { + "epoch": 0.2460169542475801, + "grad_norm": 2.0606521651861165, + "learning_rate": 3.530307103666342e-06, + "loss": 1.1338, + "step": 2046 + }, + { + "epoch": 0.24613719713821922, + "grad_norm": 2.4834271428185173, + "learning_rate": 3.5298054483926658e-06, + "loss": 1.0418, + "step": 2047 + }, + { + "epoch": 0.2462574400288583, + "grad_norm": 2.203455257687761, + "learning_rate": 3.5293035610487187e-06, + "loss": 1.0693, + "step": 2048 + }, + { + "epoch": 0.24637768291949738, + "grad_norm": 0.7270369672519728, + "learning_rate": 3.5288014417106374e-06, + "loss": 0.8573, + "step": 2049 + }, + { + "epoch": 0.24649792581013646, + "grad_norm": 2.1375560061607453, + "learning_rate": 3.528299090454593e-06, + "loss": 0.9949, + "step": 2050 + }, + { + "epoch": 0.24661816870077558, + "grad_norm": 2.415746798781941, + "learning_rate": 3.527796507356792e-06, + "loss": 1.0713, + "step": 2051 + }, + { + "epoch": 0.24673841159141466, + "grad_norm": 3.1132060464274134, + "learning_rate": 3.527293692493475e-06, + "loss": 1.1451, + "step": 2052 + }, + { + "epoch": 0.24685865448205374, + "grad_norm": 4.244141748518604, + "learning_rate": 3.52679064594092e-06, + "loss": 0.9695, + "step": 2053 + }, + { + "epoch": 0.24697889737269285, + "grad_norm": 2.369614971126989, + "learning_rate": 3.5262873677754375e-06, + "loss": 0.9877, + "step": 2054 + }, + { + "epoch": 0.24709914026333193, + "grad_norm": 1.7964333696417745, + "learning_rate": 3.5257838580733745e-06, + "loss": 1.0454, + "step": 2055 + }, + { + "epoch": 0.24721938315397102, + "grad_norm": 3.4002400145242793, + "learning_rate": 3.5252801169111138e-06, + "loss": 1.1116, + "step": 2056 + }, + { + "epoch": 0.2473396260446101, + "grad_norm": 2.012865173190777, + "learning_rate": 3.524776144365072e-06, + "loss": 1.0357, + "step": 2057 + }, + { + "epoch": 0.2474598689352492, + "grad_norm": 1.6947734344883496, + "learning_rate": 3.5242719405117016e-06, + "loss": 1.0323, + "step": 2058 + }, + { + "epoch": 0.2475801118258883, + "grad_norm": 3.1895151687072403, + "learning_rate": 3.5237675054274893e-06, + "loss": 0.9926, + "step": 2059 + }, + { + "epoch": 0.24770035471652738, + "grad_norm": 2.485096952515801, + "learning_rate": 3.5232628391889584e-06, + "loss": 1.0397, + "step": 2060 + }, + { + "epoch": 0.2478205976071665, + "grad_norm": 2.8711611785236864, + "learning_rate": 3.522757941872666e-06, + "loss": 0.8831, + "step": 2061 + }, + { + "epoch": 0.24794084049780557, + "grad_norm": 19.81787729546348, + "learning_rate": 3.5222528135552042e-06, + "loss": 1.0654, + "step": 2062 + }, + { + "epoch": 0.24806108338844465, + "grad_norm": 1.997272180926618, + "learning_rate": 3.521747454313201e-06, + "loss": 1.0428, + "step": 2063 + }, + { + "epoch": 0.24818132627908374, + "grad_norm": 3.85865931496736, + "learning_rate": 3.521241864223319e-06, + "loss": 0.908, + "step": 2064 + }, + { + "epoch": 0.24830156916972285, + "grad_norm": 0.8266915204429213, + "learning_rate": 3.5207360433622552e-06, + "loss": 0.8571, + "step": 2065 + }, + { + "epoch": 0.24842181206036193, + "grad_norm": 1.8046664942720292, + "learning_rate": 3.5202299918067437e-06, + "loss": 0.9848, + "step": 2066 + }, + { + "epoch": 0.248542054951001, + "grad_norm": 2.5359900092460634, + "learning_rate": 3.519723709633551e-06, + "loss": 0.9296, + "step": 2067 + }, + { + "epoch": 0.24866229784164012, + "grad_norm": 3.092575775180284, + "learning_rate": 3.519217196919479e-06, + "loss": 1.0746, + "step": 2068 + }, + { + "epoch": 0.2487825407322792, + "grad_norm": 1.8846634120600454, + "learning_rate": 3.518710453741367e-06, + "loss": 0.9685, + "step": 2069 + }, + { + "epoch": 0.2489027836229183, + "grad_norm": 2.7440566080610305, + "learning_rate": 3.518203480176086e-06, + "loss": 0.9117, + "step": 2070 + }, + { + "epoch": 0.2490230265135574, + "grad_norm": 1.813804945951468, + "learning_rate": 3.517696276300545e-06, + "loss": 1.0465, + "step": 2071 + }, + { + "epoch": 0.24914326940419648, + "grad_norm": 2.5114672877562, + "learning_rate": 3.517188842191685e-06, + "loss": 0.9362, + "step": 2072 + }, + { + "epoch": 0.24926351229483557, + "grad_norm": 1.8770945611830334, + "learning_rate": 3.5166811779264837e-06, + "loss": 0.9786, + "step": 2073 + }, + { + "epoch": 0.24938375518547465, + "grad_norm": 2.33858888121069, + "learning_rate": 3.5161732835819545e-06, + "loss": 1.0198, + "step": 2074 + }, + { + "epoch": 0.24950399807611376, + "grad_norm": 2.034202425386001, + "learning_rate": 3.515665159235143e-06, + "loss": 1.0701, + "step": 2075 + }, + { + "epoch": 0.24962424096675284, + "grad_norm": 2.6398534606953685, + "learning_rate": 3.5151568049631318e-06, + "loss": 0.9843, + "step": 2076 + }, + { + "epoch": 0.24974448385739192, + "grad_norm": 1.7844997008353614, + "learning_rate": 3.5146482208430385e-06, + "loss": 1.0461, + "step": 2077 + }, + { + "epoch": 0.24986472674803104, + "grad_norm": 3.0770751402032794, + "learning_rate": 3.514139406952014e-06, + "loss": 0.9157, + "step": 2078 + }, + { + "epoch": 0.24998496963867012, + "grad_norm": 1.8242733802212023, + "learning_rate": 3.5136303633672454e-06, + "loss": 1.0752, + "step": 2079 + }, + { + "epoch": 0.25010521252930923, + "grad_norm": 1.950757842864927, + "learning_rate": 3.5131210901659544e-06, + "loss": 0.9844, + "step": 2080 + }, + { + "epoch": 0.2502254554199483, + "grad_norm": 2.5744725103727637, + "learning_rate": 3.5126115874253967e-06, + "loss": 1.0574, + "step": 2081 + }, + { + "epoch": 0.2503456983105874, + "grad_norm": 2.1060085984622368, + "learning_rate": 3.5121018552228644e-06, + "loss": 1.0495, + "step": 2082 + }, + { + "epoch": 0.2504659412012265, + "grad_norm": 3.59308571688348, + "learning_rate": 3.5115918936356827e-06, + "loss": 1.0053, + "step": 2083 + }, + { + "epoch": 0.25058618409186556, + "grad_norm": 2.4147614390063485, + "learning_rate": 3.5110817027412123e-06, + "loss": 1.0289, + "step": 2084 + }, + { + "epoch": 0.25070642698250467, + "grad_norm": 2.017715789145692, + "learning_rate": 3.5105712826168493e-06, + "loss": 0.9278, + "step": 2085 + }, + { + "epoch": 0.2508266698731437, + "grad_norm": 2.000635880473759, + "learning_rate": 3.5100606333400235e-06, + "loss": 0.9451, + "step": 2086 + }, + { + "epoch": 0.25094691276378284, + "grad_norm": 2.8147136877346686, + "learning_rate": 3.5095497549882006e-06, + "loss": 1.0097, + "step": 2087 + }, + { + "epoch": 0.25106715565442195, + "grad_norm": 2.328212147560147, + "learning_rate": 3.50903864763888e-06, + "loss": 0.9634, + "step": 2088 + }, + { + "epoch": 0.251187398545061, + "grad_norm": 2.149667949318657, + "learning_rate": 3.5085273113695965e-06, + "loss": 1.0031, + "step": 2089 + }, + { + "epoch": 0.2513076414357001, + "grad_norm": 2.255687201729149, + "learning_rate": 3.508015746257919e-06, + "loss": 1.0258, + "step": 2090 + }, + { + "epoch": 0.2514278843263392, + "grad_norm": 2.6274917220575977, + "learning_rate": 3.5075039523814518e-06, + "loss": 1.0698, + "step": 2091 + }, + { + "epoch": 0.2515481272169783, + "grad_norm": 2.1850407536806635, + "learning_rate": 3.506991929817834e-06, + "loss": 1.0578, + "step": 2092 + }, + { + "epoch": 0.2516683701076174, + "grad_norm": 1.7956084722069636, + "learning_rate": 3.506479678644738e-06, + "loss": 1.0653, + "step": 2093 + }, + { + "epoch": 0.2517886129982565, + "grad_norm": 2.5859953997719334, + "learning_rate": 3.505967198939873e-06, + "loss": 0.9774, + "step": 2094 + }, + { + "epoch": 0.25190885588889556, + "grad_norm": 1.9422779850892367, + "learning_rate": 3.5054544907809813e-06, + "loss": 1.0148, + "step": 2095 + }, + { + "epoch": 0.25202909877953467, + "grad_norm": 2.1570765804513954, + "learning_rate": 3.50494155424584e-06, + "loss": 1.0405, + "step": 2096 + }, + { + "epoch": 0.2521493416701738, + "grad_norm": 1.8283948780621493, + "learning_rate": 3.504428389412262e-06, + "loss": 1.0676, + "step": 2097 + }, + { + "epoch": 0.25226958456081283, + "grad_norm": 2.3802133862019557, + "learning_rate": 3.5039149963580927e-06, + "loss": 0.9688, + "step": 2098 + }, + { + "epoch": 0.25238982745145194, + "grad_norm": 2.4992459113555126, + "learning_rate": 3.503401375161215e-06, + "loss": 0.9397, + "step": 2099 + }, + { + "epoch": 0.252510070342091, + "grad_norm": 2.148880458597406, + "learning_rate": 3.502887525899544e-06, + "loss": 1.0779, + "step": 2100 + }, + { + "epoch": 0.2526303132327301, + "grad_norm": 1.885963936583299, + "learning_rate": 3.50237344865103e-06, + "loss": 1.0615, + "step": 2101 + }, + { + "epoch": 0.2527505561233692, + "grad_norm": 2.405190271634935, + "learning_rate": 3.501859143493658e-06, + "loss": 1.0013, + "step": 2102 + }, + { + "epoch": 0.2528707990140083, + "grad_norm": 0.9457576113623237, + "learning_rate": 3.5013446105054488e-06, + "loss": 0.8434, + "step": 2103 + }, + { + "epoch": 0.2529910419046474, + "grad_norm": 1.9934495107525436, + "learning_rate": 3.5008298497644555e-06, + "loss": 0.9851, + "step": 2104 + }, + { + "epoch": 0.2531112847952865, + "grad_norm": 1.9632725889634988, + "learning_rate": 3.500314861348767e-06, + "loss": 1.1167, + "step": 2105 + }, + { + "epoch": 0.25323152768592555, + "grad_norm": 2.236318423320932, + "learning_rate": 3.499799645336507e-06, + "loss": 1.0096, + "step": 2106 + }, + { + "epoch": 0.25335177057656466, + "grad_norm": 1.5221373658919655, + "learning_rate": 3.4992842018058336e-06, + "loss": 1.1074, + "step": 2107 + }, + { + "epoch": 0.25347201346720377, + "grad_norm": 2.228202777805143, + "learning_rate": 3.4987685308349384e-06, + "loss": 1.1214, + "step": 2108 + }, + { + "epoch": 0.2535922563578428, + "grad_norm": 5.578721129462165, + "learning_rate": 3.4982526325020497e-06, + "loss": 0.852, + "step": 2109 + }, + { + "epoch": 0.25371249924848194, + "grad_norm": 2.615534565044707, + "learning_rate": 3.4977365068854273e-06, + "loss": 1.0564, + "step": 2110 + }, + { + "epoch": 0.25383274213912105, + "grad_norm": 1.848731601048363, + "learning_rate": 3.4972201540633676e-06, + "loss": 0.9718, + "step": 2111 + }, + { + "epoch": 0.2539529850297601, + "grad_norm": 1.9425039584439865, + "learning_rate": 3.4967035741142008e-06, + "loss": 1.0898, + "step": 2112 + }, + { + "epoch": 0.2540732279203992, + "grad_norm": 1.9068016754826427, + "learning_rate": 3.4961867671162917e-06, + "loss": 1.0594, + "step": 2113 + }, + { + "epoch": 0.2541934708110383, + "grad_norm": 2.391791663630663, + "learning_rate": 3.4956697331480402e-06, + "loss": 1.0112, + "step": 2114 + }, + { + "epoch": 0.2543137137016774, + "grad_norm": 1.981895459567845, + "learning_rate": 3.495152472287879e-06, + "loss": 1.0386, + "step": 2115 + }, + { + "epoch": 0.2544339565923165, + "grad_norm": 1.937722101094622, + "learning_rate": 3.4946349846142766e-06, + "loss": 0.9749, + "step": 2116 + }, + { + "epoch": 0.25455419948295555, + "grad_norm": 2.2117728552471223, + "learning_rate": 3.4941172702057353e-06, + "loss": 0.9908, + "step": 2117 + }, + { + "epoch": 0.25467444237359466, + "grad_norm": 1.94903600290944, + "learning_rate": 3.4935993291407924e-06, + "loss": 1.0436, + "step": 2118 + }, + { + "epoch": 0.25479468526423377, + "grad_norm": 2.36554593114098, + "learning_rate": 3.4930811614980183e-06, + "loss": 0.9433, + "step": 2119 + }, + { + "epoch": 0.2549149281548728, + "grad_norm": 2.405678916638144, + "learning_rate": 3.4925627673560198e-06, + "loss": 1.0284, + "step": 2120 + }, + { + "epoch": 0.25503517104551193, + "grad_norm": 2.060103092947185, + "learning_rate": 3.4920441467934357e-06, + "loss": 1.1206, + "step": 2121 + }, + { + "epoch": 0.25515541393615104, + "grad_norm": 2.416617513573296, + "learning_rate": 3.491525299888941e-06, + "loss": 1.0716, + "step": 2122 + }, + { + "epoch": 0.2552756568267901, + "grad_norm": 0.9397844773118111, + "learning_rate": 3.491006226721244e-06, + "loss": 0.8645, + "step": 2123 + }, + { + "epoch": 0.2553958997174292, + "grad_norm": 2.1120188780287696, + "learning_rate": 3.4904869273690882e-06, + "loss": 1.0086, + "step": 2124 + }, + { + "epoch": 0.2555161426080683, + "grad_norm": 2.064561333973758, + "learning_rate": 3.489967401911251e-06, + "loss": 1.1272, + "step": 2125 + }, + { + "epoch": 0.2556363854987074, + "grad_norm": 3.1045656054546145, + "learning_rate": 3.4894476504265428e-06, + "loss": 0.9327, + "step": 2126 + }, + { + "epoch": 0.2557566283893465, + "grad_norm": 0.75674311899334, + "learning_rate": 3.4889276729938104e-06, + "loss": 0.7814, + "step": 2127 + }, + { + "epoch": 0.2558768712799856, + "grad_norm": 2.6618361133422948, + "learning_rate": 3.488407469691934e-06, + "loss": 1.0435, + "step": 2128 + }, + { + "epoch": 0.25599711417062465, + "grad_norm": 2.0466520464661793, + "learning_rate": 3.487887040599828e-06, + "loss": 1.0486, + "step": 2129 + }, + { + "epoch": 0.25611735706126376, + "grad_norm": 3.387230224002063, + "learning_rate": 3.4873663857964407e-06, + "loss": 0.9976, + "step": 2130 + }, + { + "epoch": 0.2562375999519028, + "grad_norm": 2.3682333269509885, + "learning_rate": 3.4868455053607556e-06, + "loss": 0.9035, + "step": 2131 + }, + { + "epoch": 0.2563578428425419, + "grad_norm": 2.178579397432641, + "learning_rate": 3.486324399371789e-06, + "loss": 0.96, + "step": 2132 + }, + { + "epoch": 0.25647808573318104, + "grad_norm": 2.282933489278057, + "learning_rate": 3.485803067908593e-06, + "loss": 1.0282, + "step": 2133 + }, + { + "epoch": 0.2565983286238201, + "grad_norm": 2.23681901766315, + "learning_rate": 3.485281511050253e-06, + "loss": 1.0347, + "step": 2134 + }, + { + "epoch": 0.2567185715144592, + "grad_norm": 2.852628492960073, + "learning_rate": 3.484759728875889e-06, + "loss": 1.142, + "step": 2135 + }, + { + "epoch": 0.2568388144050983, + "grad_norm": 1.7402031378720044, + "learning_rate": 3.4842377214646543e-06, + "loss": 1.0481, + "step": 2136 + }, + { + "epoch": 0.25695905729573737, + "grad_norm": 1.6754986006669155, + "learning_rate": 3.483715488895737e-06, + "loss": 0.9064, + "step": 2137 + }, + { + "epoch": 0.2570793001863765, + "grad_norm": 1.9026390018889, + "learning_rate": 3.48319303124836e-06, + "loss": 1.0208, + "step": 2138 + }, + { + "epoch": 0.2571995430770156, + "grad_norm": 2.0592258937731174, + "learning_rate": 3.4826703486017798e-06, + "loss": 0.9025, + "step": 2139 + }, + { + "epoch": 0.25731978596765465, + "grad_norm": 1.813857392405514, + "learning_rate": 3.4821474410352867e-06, + "loss": 1.0044, + "step": 2140 + }, + { + "epoch": 0.25744002885829376, + "grad_norm": 0.8980437701396856, + "learning_rate": 3.481624308628205e-06, + "loss": 0.8643, + "step": 2141 + }, + { + "epoch": 0.25756027174893287, + "grad_norm": 3.1365040934879707, + "learning_rate": 3.481100951459893e-06, + "loss": 1.2371, + "step": 2142 + }, + { + "epoch": 0.2576805146395719, + "grad_norm": 1.6511197798709454, + "learning_rate": 3.4805773696097453e-06, + "loss": 1.0276, + "step": 2143 + }, + { + "epoch": 0.25780075753021103, + "grad_norm": 3.1147353391436257, + "learning_rate": 3.4800535631571874e-06, + "loss": 1.1146, + "step": 2144 + }, + { + "epoch": 0.25792100042085014, + "grad_norm": 7.532357921861992, + "learning_rate": 3.4795295321816804e-06, + "loss": 1.0011, + "step": 2145 + }, + { + "epoch": 0.2580412433114892, + "grad_norm": 2.0692648486122778, + "learning_rate": 3.47900527676272e-06, + "loss": 1.1501, + "step": 2146 + }, + { + "epoch": 0.2581614862021283, + "grad_norm": 2.010218147662135, + "learning_rate": 3.478480796979835e-06, + "loss": 1.1265, + "step": 2147 + }, + { + "epoch": 0.25828172909276736, + "grad_norm": 4.008527249558024, + "learning_rate": 3.4779560929125894e-06, + "loss": 1.0156, + "step": 2148 + }, + { + "epoch": 0.2584019719834065, + "grad_norm": 0.6678502384816817, + "learning_rate": 3.4774311646405783e-06, + "loss": 0.807, + "step": 2149 + }, + { + "epoch": 0.2585222148740456, + "grad_norm": 2.6581778318391627, + "learning_rate": 3.476906012243435e-06, + "loss": 1.0725, + "step": 2150 + }, + { + "epoch": 0.25864245776468464, + "grad_norm": 1.6900257000295156, + "learning_rate": 3.476380635800824e-06, + "loss": 1.0503, + "step": 2151 + }, + { + "epoch": 0.25876270065532375, + "grad_norm": 4.078421980176729, + "learning_rate": 3.475855035392444e-06, + "loss": 1.0973, + "step": 2152 + }, + { + "epoch": 0.25888294354596286, + "grad_norm": 2.3405244120641036, + "learning_rate": 3.475329211098029e-06, + "loss": 0.9527, + "step": 2153 + }, + { + "epoch": 0.2590031864366019, + "grad_norm": 1.7036297951245334, + "learning_rate": 3.4748031629973453e-06, + "loss": 1.0619, + "step": 2154 + }, + { + "epoch": 0.25912342932724103, + "grad_norm": 0.9086506387050783, + "learning_rate": 3.4742768911701944e-06, + "loss": 0.803, + "step": 2155 + }, + { + "epoch": 0.25924367221788014, + "grad_norm": 6.762106881552454, + "learning_rate": 3.4737503956964113e-06, + "loss": 0.9431, + "step": 2156 + }, + { + "epoch": 0.2593639151085192, + "grad_norm": 2.1349490018020125, + "learning_rate": 3.473223676655865e-06, + "loss": 0.915, + "step": 2157 + }, + { + "epoch": 0.2594841579991583, + "grad_norm": 2.1266992170459686, + "learning_rate": 3.472696734128459e-06, + "loss": 1.0398, + "step": 2158 + }, + { + "epoch": 0.2596044008897974, + "grad_norm": 3.139397528368379, + "learning_rate": 3.4721695681941286e-06, + "loss": 0.9964, + "step": 2159 + }, + { + "epoch": 0.25972464378043647, + "grad_norm": 2.7172284308011, + "learning_rate": 3.471642178932845e-06, + "loss": 1.0615, + "step": 2160 + }, + { + "epoch": 0.2598448866710756, + "grad_norm": 2.1713575666155007, + "learning_rate": 3.471114566424613e-06, + "loss": 1.1334, + "step": 2161 + }, + { + "epoch": 0.25996512956171464, + "grad_norm": 2.055169458757945, + "learning_rate": 3.4705867307494715e-06, + "loss": 0.9978, + "step": 2162 + }, + { + "epoch": 0.26008537245235375, + "grad_norm": 2.601060515307318, + "learning_rate": 3.470058671987492e-06, + "loss": 1.0864, + "step": 2163 + }, + { + "epoch": 0.26020561534299286, + "grad_norm": 1.8737937862072123, + "learning_rate": 3.4695303902187805e-06, + "loss": 1.0827, + "step": 2164 + }, + { + "epoch": 0.2603258582336319, + "grad_norm": 2.321750182625698, + "learning_rate": 3.469001885523478e-06, + "loss": 1.0237, + "step": 2165 + }, + { + "epoch": 0.260446101124271, + "grad_norm": 2.2885153054577128, + "learning_rate": 3.4684731579817568e-06, + "loss": 1.0493, + "step": 2166 + }, + { + "epoch": 0.26056634401491013, + "grad_norm": 1.6252064206991037, + "learning_rate": 3.4679442076738247e-06, + "loss": 1.0043, + "step": 2167 + }, + { + "epoch": 0.2606865869055492, + "grad_norm": 1.760046933263613, + "learning_rate": 3.4674150346799245e-06, + "loss": 1.0747, + "step": 2168 + }, + { + "epoch": 0.2608068297961883, + "grad_norm": 2.4849419096944025, + "learning_rate": 3.4668856390803295e-06, + "loss": 1.045, + "step": 2169 + }, + { + "epoch": 0.2609270726868274, + "grad_norm": 2.2186627438003144, + "learning_rate": 3.4663560209553495e-06, + "loss": 1.1324, + "step": 2170 + }, + { + "epoch": 0.26104731557746647, + "grad_norm": 2.010656339230985, + "learning_rate": 3.4658261803853267e-06, + "loss": 1.0294, + "step": 2171 + }, + { + "epoch": 0.2611675584681056, + "grad_norm": 2.1793587823017693, + "learning_rate": 3.4652961174506383e-06, + "loss": 1.0453, + "step": 2172 + }, + { + "epoch": 0.2612878013587447, + "grad_norm": 0.9673802937028712, + "learning_rate": 3.464765832231694e-06, + "loss": 0.8184, + "step": 2173 + }, + { + "epoch": 0.26140804424938374, + "grad_norm": 1.7627551631329796, + "learning_rate": 3.4642353248089373e-06, + "loss": 0.9453, + "step": 2174 + }, + { + "epoch": 0.26152828714002285, + "grad_norm": 2.5527478241803325, + "learning_rate": 3.463704595262846e-06, + "loss": 1.0408, + "step": 2175 + }, + { + "epoch": 0.26164853003066196, + "grad_norm": 2.3244983665179944, + "learning_rate": 3.463173643673931e-06, + "loss": 0.9474, + "step": 2176 + }, + { + "epoch": 0.261768772921301, + "grad_norm": 0.9053911216972135, + "learning_rate": 3.4626424701227387e-06, + "loss": 0.8729, + "step": 2177 + }, + { + "epoch": 0.26188901581194013, + "grad_norm": 0.8369143117683956, + "learning_rate": 3.4621110746898452e-06, + "loss": 0.7961, + "step": 2178 + }, + { + "epoch": 0.2620092587025792, + "grad_norm": 1.692075883137711, + "learning_rate": 3.4615794574558654e-06, + "loss": 0.988, + "step": 2179 + }, + { + "epoch": 0.2621295015932183, + "grad_norm": 2.4738817597969995, + "learning_rate": 3.4610476185014436e-06, + "loss": 1.0791, + "step": 2180 + }, + { + "epoch": 0.2622497444838574, + "grad_norm": 1.7898472559880818, + "learning_rate": 3.4605155579072597e-06, + "loss": 1.032, + "step": 2181 + }, + { + "epoch": 0.26236998737449646, + "grad_norm": 1.8844455585586406, + "learning_rate": 3.459983275754027e-06, + "loss": 0.9522, + "step": 2182 + }, + { + "epoch": 0.26249023026513557, + "grad_norm": 2.7633756121571214, + "learning_rate": 3.4594507721224918e-06, + "loss": 1.0301, + "step": 2183 + }, + { + "epoch": 0.2626104731557747, + "grad_norm": 1.6624259960941492, + "learning_rate": 3.4589180470934353e-06, + "loss": 1.0538, + "step": 2184 + }, + { + "epoch": 0.26273071604641374, + "grad_norm": 1.7444865704908143, + "learning_rate": 3.4583851007476713e-06, + "loss": 1.013, + "step": 2185 + }, + { + "epoch": 0.26285095893705285, + "grad_norm": 15.006514596890234, + "learning_rate": 3.4578519331660464e-06, + "loss": 0.9234, + "step": 2186 + }, + { + "epoch": 0.26297120182769196, + "grad_norm": 1.8994829101637183, + "learning_rate": 3.4573185444294426e-06, + "loss": 1.0592, + "step": 2187 + }, + { + "epoch": 0.263091444718331, + "grad_norm": 1.8151908499891012, + "learning_rate": 3.456784934618774e-06, + "loss": 1.0271, + "step": 2188 + }, + { + "epoch": 0.2632116876089701, + "grad_norm": 2.054865956262249, + "learning_rate": 3.4562511038149897e-06, + "loss": 1.042, + "step": 2189 + }, + { + "epoch": 0.26333193049960923, + "grad_norm": 0.8680572959034746, + "learning_rate": 3.4557170520990705e-06, + "loss": 0.815, + "step": 2190 + }, + { + "epoch": 0.2634521733902483, + "grad_norm": 1.652966825998133, + "learning_rate": 3.4551827795520324e-06, + "loss": 1.1041, + "step": 2191 + }, + { + "epoch": 0.2635724162808874, + "grad_norm": 1.8560100849921726, + "learning_rate": 3.4546482862549226e-06, + "loss": 1.0849, + "step": 2192 + }, + { + "epoch": 0.2636926591715265, + "grad_norm": 2.245264580307518, + "learning_rate": 3.4541135722888253e-06, + "loss": 1.0268, + "step": 2193 + }, + { + "epoch": 0.26381290206216557, + "grad_norm": 2.3240448702356002, + "learning_rate": 3.453578637734854e-06, + "loss": 1.0407, + "step": 2194 + }, + { + "epoch": 0.2639331449528047, + "grad_norm": 1.7386290470180135, + "learning_rate": 3.4530434826741605e-06, + "loss": 1.0244, + "step": 2195 + }, + { + "epoch": 0.26405338784344373, + "grad_norm": 1.916738329114369, + "learning_rate": 3.452508107187926e-06, + "loss": 0.9297, + "step": 2196 + }, + { + "epoch": 0.26417363073408284, + "grad_norm": 1.9092974458168595, + "learning_rate": 3.451972511357366e-06, + "loss": 1.0074, + "step": 2197 + }, + { + "epoch": 0.26429387362472195, + "grad_norm": 2.0311954197519833, + "learning_rate": 3.45143669526373e-06, + "loss": 1.0925, + "step": 2198 + }, + { + "epoch": 0.264414116515361, + "grad_norm": 0.7961946462796672, + "learning_rate": 3.450900658988302e-06, + "loss": 0.8699, + "step": 2199 + }, + { + "epoch": 0.2645343594060001, + "grad_norm": 2.2120215248798965, + "learning_rate": 3.450364402612397e-06, + "loss": 1.0158, + "step": 2200 + }, + { + "epoch": 0.26465460229663923, + "grad_norm": 1.957039168716826, + "learning_rate": 3.449827926217366e-06, + "loss": 1.0771, + "step": 2201 + }, + { + "epoch": 0.2647748451872783, + "grad_norm": 5.306007456993157, + "learning_rate": 3.449291229884591e-06, + "loss": 1.0459, + "step": 2202 + }, + { + "epoch": 0.2648950880779174, + "grad_norm": 2.00850597235789, + "learning_rate": 3.4487543136954887e-06, + "loss": 1.1042, + "step": 2203 + }, + { + "epoch": 0.2650153309685565, + "grad_norm": 1.7262542587548164, + "learning_rate": 3.448217177731509e-06, + "loss": 1.1511, + "step": 2204 + }, + { + "epoch": 0.26513557385919556, + "grad_norm": 2.0491223992148715, + "learning_rate": 3.4476798220741348e-06, + "loss": 1.015, + "step": 2205 + }, + { + "epoch": 0.26525581674983467, + "grad_norm": 1.6674961251323923, + "learning_rate": 3.4471422468048826e-06, + "loss": 1.017, + "step": 2206 + }, + { + "epoch": 0.2653760596404738, + "grad_norm": 2.600790923201728, + "learning_rate": 3.4466044520053022e-06, + "loss": 0.9667, + "step": 2207 + }, + { + "epoch": 0.26549630253111284, + "grad_norm": 1.9771136695241471, + "learning_rate": 3.446066437756977e-06, + "loss": 0.8407, + "step": 2208 + }, + { + "epoch": 0.26561654542175195, + "grad_norm": 2.148771835668905, + "learning_rate": 3.4455282041415224e-06, + "loss": 0.9888, + "step": 2209 + }, + { + "epoch": 0.265736788312391, + "grad_norm": 2.5450494694726444, + "learning_rate": 3.4449897512405894e-06, + "loss": 1.1128, + "step": 2210 + }, + { + "epoch": 0.2658570312030301, + "grad_norm": 5.475068023119721, + "learning_rate": 3.444451079135859e-06, + "loss": 0.9912, + "step": 2211 + }, + { + "epoch": 0.2659772740936692, + "grad_norm": 2.605654319548815, + "learning_rate": 3.4439121879090493e-06, + "loss": 0.9787, + "step": 2212 + }, + { + "epoch": 0.2660975169843083, + "grad_norm": 1.984111978716346, + "learning_rate": 3.4433730776419082e-06, + "loss": 1.0734, + "step": 2213 + }, + { + "epoch": 0.2662177598749474, + "grad_norm": 4.733927387694375, + "learning_rate": 3.4428337484162183e-06, + "loss": 1.0411, + "step": 2214 + }, + { + "epoch": 0.2663380027655865, + "grad_norm": 1.8864258090550097, + "learning_rate": 3.442294200313797e-06, + "loss": 1.0816, + "step": 2215 + }, + { + "epoch": 0.26645824565622556, + "grad_norm": 0.7726219192761921, + "learning_rate": 3.4417544334164916e-06, + "loss": 0.7655, + "step": 2216 + }, + { + "epoch": 0.26657848854686467, + "grad_norm": 1.6902625465734242, + "learning_rate": 3.4412144478061854e-06, + "loss": 1.0158, + "step": 2217 + }, + { + "epoch": 0.2666987314375038, + "grad_norm": 2.020467338322565, + "learning_rate": 3.4406742435647925e-06, + "loss": 0.9907, + "step": 2218 + }, + { + "epoch": 0.26681897432814283, + "grad_norm": 2.085821827170803, + "learning_rate": 3.440133820774263e-06, + "loss": 1.0279, + "step": 2219 + }, + { + "epoch": 0.26693921721878194, + "grad_norm": 2.130149436329233, + "learning_rate": 3.439593179516578e-06, + "loss": 1.0583, + "step": 2220 + }, + { + "epoch": 0.26705946010942105, + "grad_norm": 1.7648946551259133, + "learning_rate": 3.4390523198737524e-06, + "loss": 1.0459, + "step": 2221 + }, + { + "epoch": 0.2671797030000601, + "grad_norm": 1.6495549179652742, + "learning_rate": 3.4385112419278333e-06, + "loss": 0.9781, + "step": 2222 + }, + { + "epoch": 0.2672999458906992, + "grad_norm": 0.7910026266654828, + "learning_rate": 3.4379699457609033e-06, + "loss": 0.8856, + "step": 2223 + }, + { + "epoch": 0.26742018878133833, + "grad_norm": 2.1007035212325067, + "learning_rate": 3.4374284314550755e-06, + "loss": 1.1386, + "step": 2224 + }, + { + "epoch": 0.2675404316719774, + "grad_norm": 2.1692867087660637, + "learning_rate": 3.436886699092498e-06, + "loss": 1.0459, + "step": 2225 + }, + { + "epoch": 0.2676606745626165, + "grad_norm": 3.749925994142798, + "learning_rate": 3.4363447487553502e-06, + "loss": 0.9572, + "step": 2226 + }, + { + "epoch": 0.26778091745325555, + "grad_norm": 2.9661331690025863, + "learning_rate": 3.4358025805258455e-06, + "loss": 1.0197, + "step": 2227 + }, + { + "epoch": 0.26790116034389466, + "grad_norm": 1.961348653905086, + "learning_rate": 3.435260194486232e-06, + "loss": 1.0722, + "step": 2228 + }, + { + "epoch": 0.2680214032345338, + "grad_norm": 2.5416019827527894, + "learning_rate": 3.4347175907187875e-06, + "loss": 1.0565, + "step": 2229 + }, + { + "epoch": 0.26814164612517283, + "grad_norm": 3.9257056992885695, + "learning_rate": 3.4341747693058254e-06, + "loss": 1.1181, + "step": 2230 + }, + { + "epoch": 0.26826188901581194, + "grad_norm": 1.8244775102545947, + "learning_rate": 3.4336317303296916e-06, + "loss": 1.0058, + "step": 2231 + }, + { + "epoch": 0.26838213190645105, + "grad_norm": 12.12760346767409, + "learning_rate": 3.4330884738727635e-06, + "loss": 0.9908, + "step": 2232 + }, + { + "epoch": 0.2685023747970901, + "grad_norm": 2.384347390031259, + "learning_rate": 3.4325450000174535e-06, + "loss": 0.9447, + "step": 2233 + }, + { + "epoch": 0.2686226176877292, + "grad_norm": 2.35378051786649, + "learning_rate": 3.4320013088462067e-06, + "loss": 0.98, + "step": 2234 + }, + { + "epoch": 0.2687428605783683, + "grad_norm": 1.5540721960051014, + "learning_rate": 3.431457400441499e-06, + "loss": 1.0584, + "step": 2235 + }, + { + "epoch": 0.2688631034690074, + "grad_norm": 0.920010216231221, + "learning_rate": 3.4309132748858424e-06, + "loss": 0.8463, + "step": 2236 + }, + { + "epoch": 0.2689833463596465, + "grad_norm": 1.7027926339432602, + "learning_rate": 3.430368932261779e-06, + "loss": 1.0766, + "step": 2237 + }, + { + "epoch": 0.2691035892502856, + "grad_norm": 2.3476864119383682, + "learning_rate": 3.429824372651886e-06, + "loss": 0.9879, + "step": 2238 + }, + { + "epoch": 0.26922383214092466, + "grad_norm": 2.9223746622193802, + "learning_rate": 3.4292795961387732e-06, + "loss": 1.0696, + "step": 2239 + }, + { + "epoch": 0.26934407503156377, + "grad_norm": 3.142176154906716, + "learning_rate": 3.4287346028050818e-06, + "loss": 1.112, + "step": 2240 + }, + { + "epoch": 0.2694643179222028, + "grad_norm": 1.5568471505925774, + "learning_rate": 3.4281893927334866e-06, + "loss": 1.0331, + "step": 2241 + }, + { + "epoch": 0.26958456081284193, + "grad_norm": 1.8374918749363638, + "learning_rate": 3.4276439660066963e-06, + "loss": 0.9921, + "step": 2242 + }, + { + "epoch": 0.26970480370348104, + "grad_norm": 2.443351861468014, + "learning_rate": 3.427098322707452e-06, + "loss": 1.0826, + "step": 2243 + }, + { + "epoch": 0.2698250465941201, + "grad_norm": 2.1121199090976095, + "learning_rate": 3.426552462918526e-06, + "loss": 1.1305, + "step": 2244 + }, + { + "epoch": 0.2699452894847592, + "grad_norm": 2.169952749449852, + "learning_rate": 3.426006386722726e-06, + "loss": 0.97, + "step": 2245 + }, + { + "epoch": 0.2700655323753983, + "grad_norm": 2.014926996541463, + "learning_rate": 3.4254600942028914e-06, + "loss": 1.1587, + "step": 2246 + }, + { + "epoch": 0.2701857752660374, + "grad_norm": 2.1856905401633644, + "learning_rate": 3.424913585441893e-06, + "loss": 1.0605, + "step": 2247 + }, + { + "epoch": 0.2703060181566765, + "grad_norm": 2.0206315900115217, + "learning_rate": 3.4243668605226374e-06, + "loss": 1.1097, + "step": 2248 + }, + { + "epoch": 0.2704262610473156, + "grad_norm": 2.3614022343493573, + "learning_rate": 3.423819919528061e-06, + "loss": 1.064, + "step": 2249 + }, + { + "epoch": 0.27054650393795465, + "grad_norm": 2.3388235577872156, + "learning_rate": 3.4232727625411355e-06, + "loss": 1.0254, + "step": 2250 + }, + { + "epoch": 0.27066674682859376, + "grad_norm": 1.7197507545723516, + "learning_rate": 3.4227253896448626e-06, + "loss": 1.1046, + "step": 2251 + }, + { + "epoch": 0.2707869897192329, + "grad_norm": 2.435090855150862, + "learning_rate": 3.42217780092228e-06, + "loss": 1.0623, + "step": 2252 + }, + { + "epoch": 0.27090723260987193, + "grad_norm": 0.8141921103128718, + "learning_rate": 3.421629996456456e-06, + "loss": 0.8411, + "step": 2253 + }, + { + "epoch": 0.27102747550051104, + "grad_norm": 1.8549368326788813, + "learning_rate": 3.421081976330491e-06, + "loss": 1.0648, + "step": 2254 + }, + { + "epoch": 0.27114771839115015, + "grad_norm": 1.9116118644014113, + "learning_rate": 3.4205337406275207e-06, + "loss": 1.1167, + "step": 2255 + }, + { + "epoch": 0.2712679612817892, + "grad_norm": 2.2889727161492943, + "learning_rate": 3.4199852894307114e-06, + "loss": 0.99, + "step": 2256 + }, + { + "epoch": 0.2713882041724283, + "grad_norm": 5.238390848149698, + "learning_rate": 3.419436622823262e-06, + "loss": 1.0265, + "step": 2257 + }, + { + "epoch": 0.27150844706306737, + "grad_norm": 3.1378816228144086, + "learning_rate": 3.4188877408884063e-06, + "loss": 0.9828, + "step": 2258 + }, + { + "epoch": 0.2716286899537065, + "grad_norm": 2.910092346718685, + "learning_rate": 3.4183386437094088e-06, + "loss": 0.8903, + "step": 2259 + }, + { + "epoch": 0.2717489328443456, + "grad_norm": 2.54597213749649, + "learning_rate": 3.417789331369565e-06, + "loss": 1.0607, + "step": 2260 + }, + { + "epoch": 0.27186917573498465, + "grad_norm": 2.2697522684441624, + "learning_rate": 3.4172398039522088e-06, + "loss": 1.1514, + "step": 2261 + }, + { + "epoch": 0.27198941862562376, + "grad_norm": 1.764463937526541, + "learning_rate": 3.4166900615407e-06, + "loss": 1.0357, + "step": 2262 + }, + { + "epoch": 0.27210966151626287, + "grad_norm": 1.9649115184829364, + "learning_rate": 3.416140104218436e-06, + "loss": 0.9878, + "step": 2263 + }, + { + "epoch": 0.2722299044069019, + "grad_norm": 0.8500442970177466, + "learning_rate": 3.4155899320688437e-06, + "loss": 0.9347, + "step": 2264 + }, + { + "epoch": 0.27235014729754103, + "grad_norm": 2.2858031310968316, + "learning_rate": 3.415039545175384e-06, + "loss": 0.9817, + "step": 2265 + }, + { + "epoch": 0.27247039018818014, + "grad_norm": 2.1883596638443747, + "learning_rate": 3.414488943621551e-06, + "loss": 0.8897, + "step": 2266 + }, + { + "epoch": 0.2725906330788192, + "grad_norm": 2.000552339578026, + "learning_rate": 3.41393812749087e-06, + "loss": 0.9773, + "step": 2267 + }, + { + "epoch": 0.2727108759694583, + "grad_norm": 2.9188080873774385, + "learning_rate": 3.4133870968668984e-06, + "loss": 0.96, + "step": 2268 + }, + { + "epoch": 0.2728311188600974, + "grad_norm": 5.124041397880426, + "learning_rate": 3.412835851833229e-06, + "loss": 1.0221, + "step": 2269 + }, + { + "epoch": 0.2729513617507365, + "grad_norm": 1.9531552732028845, + "learning_rate": 3.4122843924734834e-06, + "loss": 1.0183, + "step": 2270 + }, + { + "epoch": 0.2730716046413756, + "grad_norm": 3.3667481881522945, + "learning_rate": 3.411732718871319e-06, + "loss": 1.1184, + "step": 2271 + }, + { + "epoch": 0.27319184753201464, + "grad_norm": 2.8170771335472837, + "learning_rate": 3.4111808311104227e-06, + "loss": 1.0282, + "step": 2272 + }, + { + "epoch": 0.27331209042265375, + "grad_norm": 2.073205732832763, + "learning_rate": 3.410628729274517e-06, + "loss": 0.9333, + "step": 2273 + }, + { + "epoch": 0.27343233331329286, + "grad_norm": 2.1045644654786577, + "learning_rate": 3.4100764134473546e-06, + "loss": 1.0661, + "step": 2274 + }, + { + "epoch": 0.2735525762039319, + "grad_norm": 2.31906433442607, + "learning_rate": 3.4095238837127215e-06, + "loss": 1.0909, + "step": 2275 + }, + { + "epoch": 0.27367281909457103, + "grad_norm": 1.9452056242604012, + "learning_rate": 3.4089711401544355e-06, + "loss": 1.0298, + "step": 2276 + }, + { + "epoch": 0.27379306198521014, + "grad_norm": 2.382121802110533, + "learning_rate": 3.4084181828563486e-06, + "loss": 0.9172, + "step": 2277 + }, + { + "epoch": 0.2739133048758492, + "grad_norm": 1.7265658141229505, + "learning_rate": 3.4078650119023428e-06, + "loss": 0.9425, + "step": 2278 + }, + { + "epoch": 0.2740335477664883, + "grad_norm": 2.1979870777392416, + "learning_rate": 3.4073116273763337e-06, + "loss": 0.9777, + "step": 2279 + }, + { + "epoch": 0.2741537906571274, + "grad_norm": 1.9587334636679143, + "learning_rate": 3.40675802936227e-06, + "loss": 1.0483, + "step": 2280 + }, + { + "epoch": 0.27427403354776647, + "grad_norm": 2.0128764256591523, + "learning_rate": 3.4062042179441318e-06, + "loss": 0.9546, + "step": 2281 + }, + { + "epoch": 0.2743942764384056, + "grad_norm": 1.83370056231321, + "learning_rate": 3.4056501932059314e-06, + "loss": 1.0446, + "step": 2282 + }, + { + "epoch": 0.2745145193290447, + "grad_norm": 0.7744234909503481, + "learning_rate": 3.405095955231715e-06, + "loss": 0.8188, + "step": 2283 + }, + { + "epoch": 0.27463476221968375, + "grad_norm": 2.2914926173844297, + "learning_rate": 3.4045415041055585e-06, + "loss": 1.1777, + "step": 2284 + }, + { + "epoch": 0.27475500511032286, + "grad_norm": 2.614394699120955, + "learning_rate": 3.4039868399115728e-06, + "loss": 1.0208, + "step": 2285 + }, + { + "epoch": 0.27487524800096197, + "grad_norm": 1.994752915074874, + "learning_rate": 3.4034319627339003e-06, + "loss": 1.0444, + "step": 2286 + }, + { + "epoch": 0.274995490891601, + "grad_norm": 2.4911436567563525, + "learning_rate": 3.402876872656715e-06, + "loss": 0.9332, + "step": 2287 + }, + { + "epoch": 0.27511573378224013, + "grad_norm": 2.0244339658105646, + "learning_rate": 3.402321569764223e-06, + "loss": 1.1369, + "step": 2288 + }, + { + "epoch": 0.2752359766728792, + "grad_norm": 1.8485586531364628, + "learning_rate": 3.4017660541406635e-06, + "loss": 1.078, + "step": 2289 + }, + { + "epoch": 0.2753562195635183, + "grad_norm": 1.7977891006684232, + "learning_rate": 3.4012103258703092e-06, + "loss": 0.9826, + "step": 2290 + }, + { + "epoch": 0.2754764624541574, + "grad_norm": 2.149379898070145, + "learning_rate": 3.4006543850374616e-06, + "loss": 1.0652, + "step": 2291 + }, + { + "epoch": 0.27559670534479647, + "grad_norm": 2.0361385500893383, + "learning_rate": 3.400098231726458e-06, + "loss": 0.9868, + "step": 2292 + }, + { + "epoch": 0.2757169482354356, + "grad_norm": 2.397446418298703, + "learning_rate": 3.3995418660216657e-06, + "loss": 1.1062, + "step": 2293 + }, + { + "epoch": 0.2758371911260747, + "grad_norm": 1.9989506829854038, + "learning_rate": 3.3989852880074848e-06, + "loss": 1.0434, + "step": 2294 + }, + { + "epoch": 0.27595743401671374, + "grad_norm": 0.7603775345300553, + "learning_rate": 3.398428497768348e-06, + "loss": 0.8444, + "step": 2295 + }, + { + "epoch": 0.27607767690735285, + "grad_norm": 4.109307930402651, + "learning_rate": 3.3978714953887205e-06, + "loss": 0.9593, + "step": 2296 + }, + { + "epoch": 0.27619791979799196, + "grad_norm": 1.8414919976074284, + "learning_rate": 3.397314280953098e-06, + "loss": 1.1036, + "step": 2297 + }, + { + "epoch": 0.276318162688631, + "grad_norm": 1.992142860062059, + "learning_rate": 3.3967568545460108e-06, + "loss": 1.0435, + "step": 2298 + }, + { + "epoch": 0.27643840557927013, + "grad_norm": 1.9342720460378617, + "learning_rate": 3.3961992162520185e-06, + "loss": 1.0433, + "step": 2299 + }, + { + "epoch": 0.27655864846990924, + "grad_norm": 2.2260998697010006, + "learning_rate": 3.3956413661557156e-06, + "loss": 0.9579, + "step": 2300 + }, + { + "epoch": 0.2766788913605483, + "grad_norm": 2.1350294056638464, + "learning_rate": 3.3950833043417273e-06, + "loss": 0.9016, + "step": 2301 + }, + { + "epoch": 0.2767991342511874, + "grad_norm": 3.331902737392019, + "learning_rate": 3.3945250308947105e-06, + "loss": 0.9686, + "step": 2302 + }, + { + "epoch": 0.2769193771418265, + "grad_norm": 1.2879401204690688, + "learning_rate": 3.3939665458993556e-06, + "loss": 0.9223, + "step": 2303 + }, + { + "epoch": 0.27703962003246557, + "grad_norm": 2.4079843191225594, + "learning_rate": 3.3934078494403843e-06, + "loss": 1.0063, + "step": 2304 + }, + { + "epoch": 0.2771598629231047, + "grad_norm": 2.198319300312006, + "learning_rate": 3.3928489416025495e-06, + "loss": 1.0504, + "step": 2305 + }, + { + "epoch": 0.27728010581374374, + "grad_norm": 2.25897208139055, + "learning_rate": 3.392289822470638e-06, + "loss": 1.0265, + "step": 2306 + }, + { + "epoch": 0.27740034870438285, + "grad_norm": 4.616604582625032, + "learning_rate": 3.3917304921294674e-06, + "loss": 0.9961, + "step": 2307 + }, + { + "epoch": 0.27752059159502196, + "grad_norm": 1.8376824061365922, + "learning_rate": 3.3911709506638876e-06, + "loss": 1.0441, + "step": 2308 + }, + { + "epoch": 0.277640834485661, + "grad_norm": 2.204116922645304, + "learning_rate": 3.390611198158781e-06, + "loss": 1.0474, + "step": 2309 + }, + { + "epoch": 0.2777610773763001, + "grad_norm": 2.715848229191739, + "learning_rate": 3.3900512346990612e-06, + "loss": 1.1371, + "step": 2310 + }, + { + "epoch": 0.27788132026693924, + "grad_norm": 1.835329515146722, + "learning_rate": 3.389491060369674e-06, + "loss": 0.8928, + "step": 2311 + }, + { + "epoch": 0.2780015631575783, + "grad_norm": 2.229248889107859, + "learning_rate": 3.388930675255598e-06, + "loss": 1.1313, + "step": 2312 + }, + { + "epoch": 0.2781218060482174, + "grad_norm": 3.0985426338122646, + "learning_rate": 3.388370079441843e-06, + "loss": 1.0317, + "step": 2313 + }, + { + "epoch": 0.2782420489388565, + "grad_norm": 4.2319817906152, + "learning_rate": 3.3878092730134505e-06, + "loss": 1.1675, + "step": 2314 + }, + { + "epoch": 0.27836229182949557, + "grad_norm": 2.205755624683744, + "learning_rate": 3.3872482560554947e-06, + "loss": 1.0467, + "step": 2315 + }, + { + "epoch": 0.2784825347201347, + "grad_norm": 0.8035503982637905, + "learning_rate": 3.386687028653082e-06, + "loss": 0.808, + "step": 2316 + }, + { + "epoch": 0.2786027776107738, + "grad_norm": 1.8037361180438136, + "learning_rate": 3.386125590891349e-06, + "loss": 1.0888, + "step": 2317 + }, + { + "epoch": 0.27872302050141284, + "grad_norm": 2.007776163566507, + "learning_rate": 3.3855639428554657e-06, + "loss": 1.0679, + "step": 2318 + }, + { + "epoch": 0.27884326339205195, + "grad_norm": 1.8799031045982724, + "learning_rate": 3.385002084630635e-06, + "loss": 1.0426, + "step": 2319 + }, + { + "epoch": 0.278963506282691, + "grad_norm": 4.475783355174383, + "learning_rate": 3.384440016302088e-06, + "loss": 1.0834, + "step": 2320 + }, + { + "epoch": 0.2790837491733301, + "grad_norm": 2.1308288292308233, + "learning_rate": 3.3838777379550923e-06, + "loss": 0.8629, + "step": 2321 + }, + { + "epoch": 0.27920399206396923, + "grad_norm": 2.3480786849554653, + "learning_rate": 3.383315249674944e-06, + "loss": 1.0252, + "step": 2322 + }, + { + "epoch": 0.2793242349546083, + "grad_norm": 2.256604567799826, + "learning_rate": 3.3827525515469715e-06, + "loss": 1.1008, + "step": 2323 + }, + { + "epoch": 0.2794444778452474, + "grad_norm": 2.2283475446378653, + "learning_rate": 3.3821896436565367e-06, + "loss": 0.9505, + "step": 2324 + }, + { + "epoch": 0.2795647207358865, + "grad_norm": 1.9078939884412789, + "learning_rate": 3.381626526089032e-06, + "loss": 0.9423, + "step": 2325 + }, + { + "epoch": 0.27968496362652556, + "grad_norm": 2.5286969154507153, + "learning_rate": 3.3810631989298815e-06, + "loss": 1.0285, + "step": 2326 + }, + { + "epoch": 0.2798052065171647, + "grad_norm": 2.046212591482971, + "learning_rate": 3.3804996622645423e-06, + "loss": 1.0786, + "step": 2327 + }, + { + "epoch": 0.2799254494078038, + "grad_norm": 2.638799273212929, + "learning_rate": 3.3799359161785015e-06, + "loss": 1.1348, + "step": 2328 + }, + { + "epoch": 0.28004569229844284, + "grad_norm": 1.9023582880925491, + "learning_rate": 3.3793719607572798e-06, + "loss": 1.0979, + "step": 2329 + }, + { + "epoch": 0.28016593518908195, + "grad_norm": 2.1062361357371264, + "learning_rate": 3.378807796086428e-06, + "loss": 1.0078, + "step": 2330 + }, + { + "epoch": 0.28028617807972106, + "grad_norm": 2.1555097110806503, + "learning_rate": 3.37824342225153e-06, + "loss": 1.0057, + "step": 2331 + }, + { + "epoch": 0.2804064209703601, + "grad_norm": 2.017394833949503, + "learning_rate": 3.3776788393382006e-06, + "loss": 1.0192, + "step": 2332 + }, + { + "epoch": 0.2805266638609992, + "grad_norm": 3.7388299521388695, + "learning_rate": 3.3771140474320872e-06, + "loss": 1.0031, + "step": 2333 + }, + { + "epoch": 0.28064690675163834, + "grad_norm": 2.0082199929995723, + "learning_rate": 3.3765490466188664e-06, + "loss": 1.0326, + "step": 2334 + }, + { + "epoch": 0.2807671496422774, + "grad_norm": 2.4963734071625345, + "learning_rate": 3.3759838369842508e-06, + "loss": 0.9712, + "step": 2335 + }, + { + "epoch": 0.2808873925329165, + "grad_norm": 1.9620279874331736, + "learning_rate": 3.375418418613981e-06, + "loss": 0.971, + "step": 2336 + }, + { + "epoch": 0.28100763542355556, + "grad_norm": 2.32353043822481, + "learning_rate": 3.374852791593831e-06, + "loss": 1.0714, + "step": 2337 + }, + { + "epoch": 0.28112787831419467, + "grad_norm": 2.878341225983431, + "learning_rate": 3.374286956009605e-06, + "loss": 0.7843, + "step": 2338 + }, + { + "epoch": 0.2812481212048338, + "grad_norm": 2.1084555423383566, + "learning_rate": 3.3737209119471405e-06, + "loss": 0.9935, + "step": 2339 + }, + { + "epoch": 0.28136836409547283, + "grad_norm": 2.3159824646267815, + "learning_rate": 3.373154659492306e-06, + "loss": 0.8765, + "step": 2340 + }, + { + "epoch": 0.28148860698611194, + "grad_norm": 1.993514154051474, + "learning_rate": 3.3725881987310016e-06, + "loss": 1.0912, + "step": 2341 + }, + { + "epoch": 0.28160884987675106, + "grad_norm": 1.98482622438117, + "learning_rate": 3.372021529749159e-06, + "loss": 1.119, + "step": 2342 + }, + { + "epoch": 0.2817290927673901, + "grad_norm": 1.8520715713842661, + "learning_rate": 3.3714546526327405e-06, + "loss": 1.1654, + "step": 2343 + }, + { + "epoch": 0.2818493356580292, + "grad_norm": 2.047837358752006, + "learning_rate": 3.3708875674677423e-06, + "loss": 1.1116, + "step": 2344 + }, + { + "epoch": 0.28196957854866833, + "grad_norm": 2.358426541305118, + "learning_rate": 3.37032027434019e-06, + "loss": 1.0803, + "step": 2345 + }, + { + "epoch": 0.2820898214393074, + "grad_norm": 1.742661621173033, + "learning_rate": 3.369752773336141e-06, + "loss": 1.0683, + "step": 2346 + }, + { + "epoch": 0.2822100643299465, + "grad_norm": 1.7773194028423391, + "learning_rate": 3.3691850645416864e-06, + "loss": 1.0236, + "step": 2347 + }, + { + "epoch": 0.2823303072205856, + "grad_norm": 2.0916649984325226, + "learning_rate": 3.368617148042945e-06, + "loss": 1.0695, + "step": 2348 + }, + { + "epoch": 0.28245055011122466, + "grad_norm": 2.0111688843865543, + "learning_rate": 3.368049023926071e-06, + "loss": 1.0803, + "step": 2349 + }, + { + "epoch": 0.2825707930018638, + "grad_norm": 1.6056279384278775, + "learning_rate": 3.3674806922772476e-06, + "loss": 1.0728, + "step": 2350 + }, + { + "epoch": 0.28269103589250283, + "grad_norm": 1.711735578460358, + "learning_rate": 3.3669121531826904e-06, + "loss": 0.9903, + "step": 2351 + }, + { + "epoch": 0.28281127878314194, + "grad_norm": 2.096337953745697, + "learning_rate": 3.366343406728647e-06, + "loss": 1.0708, + "step": 2352 + }, + { + "epoch": 0.28293152167378105, + "grad_norm": 1.7054387046694237, + "learning_rate": 3.3657744530013946e-06, + "loss": 0.9293, + "step": 2353 + }, + { + "epoch": 0.2830517645644201, + "grad_norm": 1.8682100058059472, + "learning_rate": 3.3652052920872437e-06, + "loss": 0.9511, + "step": 2354 + }, + { + "epoch": 0.2831720074550592, + "grad_norm": 2.6707483782971773, + "learning_rate": 3.3646359240725355e-06, + "loss": 1.0973, + "step": 2355 + }, + { + "epoch": 0.2832922503456983, + "grad_norm": 2.284437382755588, + "learning_rate": 3.364066349043643e-06, + "loss": 0.9142, + "step": 2356 + }, + { + "epoch": 0.2834124932363374, + "grad_norm": 1.8198675176588635, + "learning_rate": 3.363496567086969e-06, + "loss": 1.0571, + "step": 2357 + }, + { + "epoch": 0.2835327361269765, + "grad_norm": 1.9943987373892793, + "learning_rate": 3.3629265782889506e-06, + "loss": 0.9914, + "step": 2358 + }, + { + "epoch": 0.2836529790176156, + "grad_norm": 2.0240935804799842, + "learning_rate": 3.362356382736054e-06, + "loss": 0.9558, + "step": 2359 + }, + { + "epoch": 0.28377322190825466, + "grad_norm": 2.047291138258372, + "learning_rate": 3.361785980514777e-06, + "loss": 1.1462, + "step": 2360 + }, + { + "epoch": 0.28389346479889377, + "grad_norm": 3.681026919589051, + "learning_rate": 3.361215371711649e-06, + "loss": 1.0073, + "step": 2361 + }, + { + "epoch": 0.2840137076895329, + "grad_norm": 2.115543416871758, + "learning_rate": 3.3606445564132326e-06, + "loss": 1.0744, + "step": 2362 + }, + { + "epoch": 0.28413395058017193, + "grad_norm": 2.147456496556345, + "learning_rate": 3.360073534706118e-06, + "loss": 1.0634, + "step": 2363 + }, + { + "epoch": 0.28425419347081105, + "grad_norm": 3.4804381980368513, + "learning_rate": 3.35950230667693e-06, + "loss": 0.9978, + "step": 2364 + }, + { + "epoch": 0.28437443636145016, + "grad_norm": 3.1257916782837323, + "learning_rate": 3.358930872412323e-06, + "loss": 1.1005, + "step": 2365 + }, + { + "epoch": 0.2844946792520892, + "grad_norm": 1.9717367616491734, + "learning_rate": 3.3583592319989825e-06, + "loss": 1.0467, + "step": 2366 + }, + { + "epoch": 0.2846149221427283, + "grad_norm": 2.467098601215135, + "learning_rate": 3.357787385523627e-06, + "loss": 0.928, + "step": 2367 + }, + { + "epoch": 0.2847351650333674, + "grad_norm": 1.8968163621062413, + "learning_rate": 3.3572153330730048e-06, + "loss": 1.0666, + "step": 2368 + }, + { + "epoch": 0.2848554079240065, + "grad_norm": 0.8321036365523549, + "learning_rate": 3.3566430747338956e-06, + "loss": 0.885, + "step": 2369 + }, + { + "epoch": 0.2849756508146456, + "grad_norm": 16.747301482137168, + "learning_rate": 3.35607061059311e-06, + "loss": 1.1033, + "step": 2370 + }, + { + "epoch": 0.28509589370528465, + "grad_norm": 1.789087153768792, + "learning_rate": 3.3554979407374917e-06, + "loss": 0.9883, + "step": 2371 + }, + { + "epoch": 0.28521613659592376, + "grad_norm": 2.0969606521327524, + "learning_rate": 3.3549250652539134e-06, + "loss": 0.9748, + "step": 2372 + }, + { + "epoch": 0.2853363794865629, + "grad_norm": 1.8266080121324806, + "learning_rate": 3.3543519842292794e-06, + "loss": 1.0561, + "step": 2373 + }, + { + "epoch": 0.28545662237720193, + "grad_norm": 2.404925093089001, + "learning_rate": 3.353778697750527e-06, + "loss": 1.0742, + "step": 2374 + }, + { + "epoch": 0.28557686526784104, + "grad_norm": 2.587141695876168, + "learning_rate": 3.353205205904622e-06, + "loss": 1.1313, + "step": 2375 + }, + { + "epoch": 0.28569710815848015, + "grad_norm": 2.643291143182482, + "learning_rate": 3.3526315087785637e-06, + "loss": 0.9551, + "step": 2376 + }, + { + "epoch": 0.2858173510491192, + "grad_norm": 2.0222423661920623, + "learning_rate": 3.3520576064593805e-06, + "loss": 1.0446, + "step": 2377 + }, + { + "epoch": 0.2859375939397583, + "grad_norm": 1.549748920285839, + "learning_rate": 3.3514834990341337e-06, + "loss": 1.0613, + "step": 2378 + }, + { + "epoch": 0.2860578368303974, + "grad_norm": 4.76084585223674, + "learning_rate": 3.3509091865899144e-06, + "loss": 1.1682, + "step": 2379 + }, + { + "epoch": 0.2861780797210365, + "grad_norm": 1.8011807145526626, + "learning_rate": 3.350334669213846e-06, + "loss": 0.9461, + "step": 2380 + }, + { + "epoch": 0.2862983226116756, + "grad_norm": 2.000497756052892, + "learning_rate": 3.3497599469930816e-06, + "loss": 0.989, + "step": 2381 + }, + { + "epoch": 0.28641856550231465, + "grad_norm": 2.4184498839770474, + "learning_rate": 3.349185020014807e-06, + "loss": 1.0699, + "step": 2382 + }, + { + "epoch": 0.28653880839295376, + "grad_norm": 2.364758948528064, + "learning_rate": 3.348609888366237e-06, + "loss": 0.9855, + "step": 2383 + }, + { + "epoch": 0.28665905128359287, + "grad_norm": 2.108534242709718, + "learning_rate": 3.348034552134619e-06, + "loss": 0.8668, + "step": 2384 + }, + { + "epoch": 0.2867792941742319, + "grad_norm": 1.9057108632602546, + "learning_rate": 3.3474590114072316e-06, + "loss": 1.0827, + "step": 2385 + }, + { + "epoch": 0.28689953706487104, + "grad_norm": 2.0062019033400733, + "learning_rate": 3.3468832662713836e-06, + "loss": 1.0701, + "step": 2386 + }, + { + "epoch": 0.28701977995551015, + "grad_norm": 2.1555275190415593, + "learning_rate": 3.346307316814415e-06, + "loss": 1.073, + "step": 2387 + }, + { + "epoch": 0.2871400228461492, + "grad_norm": 2.230409638110918, + "learning_rate": 3.3457311631236965e-06, + "loss": 0.9972, + "step": 2388 + }, + { + "epoch": 0.2872602657367883, + "grad_norm": 1.8412226796243671, + "learning_rate": 3.345154805286631e-06, + "loss": 1.0822, + "step": 2389 + }, + { + "epoch": 0.2873805086274274, + "grad_norm": 2.175638683860153, + "learning_rate": 3.344578243390651e-06, + "loss": 1.0031, + "step": 2390 + }, + { + "epoch": 0.2875007515180665, + "grad_norm": 13.447483346322068, + "learning_rate": 3.3440014775232206e-06, + "loss": 1.0196, + "step": 2391 + }, + { + "epoch": 0.2876209944087056, + "grad_norm": 1.8632195740414785, + "learning_rate": 3.343424507771834e-06, + "loss": 0.9505, + "step": 2392 + }, + { + "epoch": 0.2877412372993447, + "grad_norm": 1.9165863213768428, + "learning_rate": 3.342847334224018e-06, + "loss": 1.1024, + "step": 2393 + }, + { + "epoch": 0.28786148018998375, + "grad_norm": 0.9763184204254169, + "learning_rate": 3.342269956967329e-06, + "loss": 0.8618, + "step": 2394 + }, + { + "epoch": 0.28798172308062286, + "grad_norm": 5.105321180083251, + "learning_rate": 3.341692376089355e-06, + "loss": 0.9616, + "step": 2395 + }, + { + "epoch": 0.288101965971262, + "grad_norm": 3.1905123370359063, + "learning_rate": 3.3411145916777146e-06, + "loss": 1.0835, + "step": 2396 + }, + { + "epoch": 0.28822220886190103, + "grad_norm": 2.2453505903714994, + "learning_rate": 3.3405366038200566e-06, + "loss": 1.1456, + "step": 2397 + }, + { + "epoch": 0.28834245175254014, + "grad_norm": 2.63414379838452, + "learning_rate": 3.3399584126040617e-06, + "loss": 1.0924, + "step": 2398 + }, + { + "epoch": 0.2884626946431792, + "grad_norm": 1.8446977410152556, + "learning_rate": 3.339380018117441e-06, + "loss": 1.1474, + "step": 2399 + }, + { + "epoch": 0.2885829375338183, + "grad_norm": 2.845576108717342, + "learning_rate": 3.3388014204479366e-06, + "loss": 1.0164, + "step": 2400 + }, + { + "epoch": 0.2887031804244574, + "grad_norm": 2.1319601850173626, + "learning_rate": 3.338222619683321e-06, + "loss": 1.1549, + "step": 2401 + }, + { + "epoch": 0.2888234233150965, + "grad_norm": 3.1659657723268912, + "learning_rate": 3.337643615911398e-06, + "loss": 0.9744, + "step": 2402 + }, + { + "epoch": 0.2889436662057356, + "grad_norm": 2.1247751453589063, + "learning_rate": 3.3370644092200026e-06, + "loss": 1.0324, + "step": 2403 + }, + { + "epoch": 0.2890639090963747, + "grad_norm": 1.8269901502706003, + "learning_rate": 3.3364849996969985e-06, + "loss": 1.0218, + "step": 2404 + }, + { + "epoch": 0.28918415198701375, + "grad_norm": 1.8451903584600309, + "learning_rate": 3.335905387430283e-06, + "loss": 1.0918, + "step": 2405 + }, + { + "epoch": 0.28930439487765286, + "grad_norm": 1.911552683262659, + "learning_rate": 3.335325572507782e-06, + "loss": 1.0659, + "step": 2406 + }, + { + "epoch": 0.28942463776829197, + "grad_norm": 1.8818466431689838, + "learning_rate": 3.3347455550174537e-06, + "loss": 0.9751, + "step": 2407 + }, + { + "epoch": 0.289544880658931, + "grad_norm": 1.9591209234719649, + "learning_rate": 3.3341653350472864e-06, + "loss": 0.9192, + "step": 2408 + }, + { + "epoch": 0.28966512354957014, + "grad_norm": 2.4871553421865293, + "learning_rate": 3.333584912685298e-06, + "loss": 0.9295, + "step": 2409 + }, + { + "epoch": 0.28978536644020925, + "grad_norm": 0.865862512883124, + "learning_rate": 3.3330042880195385e-06, + "loss": 0.7926, + "step": 2410 + }, + { + "epoch": 0.2899056093308483, + "grad_norm": 1.8282518913377148, + "learning_rate": 3.3324234611380888e-06, + "loss": 1.026, + "step": 2411 + }, + { + "epoch": 0.2900258522214874, + "grad_norm": 1.6432049039650518, + "learning_rate": 3.3318424321290596e-06, + "loss": 1.0562, + "step": 2412 + }, + { + "epoch": 0.2901460951121265, + "grad_norm": 0.829800979123994, + "learning_rate": 3.3312612010805917e-06, + "loss": 0.8374, + "step": 2413 + }, + { + "epoch": 0.2902663380027656, + "grad_norm": 1.7976423532412185, + "learning_rate": 3.330679768080858e-06, + "loss": 0.943, + "step": 2414 + }, + { + "epoch": 0.2903865808934047, + "grad_norm": 2.1611813532449724, + "learning_rate": 3.3300981332180627e-06, + "loss": 1.0729, + "step": 2415 + }, + { + "epoch": 0.29050682378404374, + "grad_norm": 1.9266025115202774, + "learning_rate": 3.3295162965804373e-06, + "loss": 1.0349, + "step": 2416 + }, + { + "epoch": 0.29062706667468285, + "grad_norm": 2.397376307219363, + "learning_rate": 3.328934258256247e-06, + "loss": 1.0269, + "step": 2417 + }, + { + "epoch": 0.29074730956532197, + "grad_norm": 2.401927781727426, + "learning_rate": 3.3283520183337856e-06, + "loss": 0.9065, + "step": 2418 + }, + { + "epoch": 0.290867552455961, + "grad_norm": 1.8374136975840079, + "learning_rate": 3.3277695769013797e-06, + "loss": 0.927, + "step": 2419 + }, + { + "epoch": 0.29098779534660013, + "grad_norm": 2.194255099061707, + "learning_rate": 3.327186934047385e-06, + "loss": 1.0113, + "step": 2420 + }, + { + "epoch": 0.29110803823723924, + "grad_norm": 2.2383665697703052, + "learning_rate": 3.3266040898601877e-06, + "loss": 0.8953, + "step": 2421 + }, + { + "epoch": 0.2912282811278783, + "grad_norm": 2.2541863384786947, + "learning_rate": 3.3260210444282045e-06, + "loss": 1.0158, + "step": 2422 + }, + { + "epoch": 0.2913485240185174, + "grad_norm": 2.0893834828288154, + "learning_rate": 3.325437797839883e-06, + "loss": 0.9668, + "step": 2423 + }, + { + "epoch": 0.2914687669091565, + "grad_norm": 2.4708832807033057, + "learning_rate": 3.3248543501837015e-06, + "loss": 0.9902, + "step": 2424 + }, + { + "epoch": 0.2915890097997956, + "grad_norm": 2.0040345030077713, + "learning_rate": 3.3242707015481684e-06, + "loss": 1.0115, + "step": 2425 + }, + { + "epoch": 0.2917092526904347, + "grad_norm": 1.9550520888508929, + "learning_rate": 3.323686852021823e-06, + "loss": 1.0424, + "step": 2426 + }, + { + "epoch": 0.2918294955810738, + "grad_norm": 3.9638566014425036, + "learning_rate": 3.323102801693235e-06, + "loss": 1.0394, + "step": 2427 + }, + { + "epoch": 0.29194973847171285, + "grad_norm": 2.247677558105789, + "learning_rate": 3.322518550651003e-06, + "loss": 1.0411, + "step": 2428 + }, + { + "epoch": 0.29206998136235196, + "grad_norm": 2.413415618445604, + "learning_rate": 3.3219340989837586e-06, + "loss": 1.0539, + "step": 2429 + }, + { + "epoch": 0.292190224252991, + "grad_norm": 1.9562657852267238, + "learning_rate": 3.3213494467801625e-06, + "loss": 1.0427, + "step": 2430 + }, + { + "epoch": 0.2923104671436301, + "grad_norm": 2.696556710813813, + "learning_rate": 3.3207645941289063e-06, + "loss": 0.9475, + "step": 2431 + }, + { + "epoch": 0.29243071003426924, + "grad_norm": 1.9031628484360545, + "learning_rate": 3.320179541118711e-06, + "loss": 1.0414, + "step": 2432 + }, + { + "epoch": 0.2925509529249083, + "grad_norm": 1.0083322884945671, + "learning_rate": 3.3195942878383293e-06, + "loss": 0.8413, + "step": 2433 + }, + { + "epoch": 0.2926711958155474, + "grad_norm": 2.25897239801938, + "learning_rate": 3.319008834376543e-06, + "loss": 1.0146, + "step": 2434 + }, + { + "epoch": 0.2927914387061865, + "grad_norm": 2.876266034880251, + "learning_rate": 3.3184231808221654e-06, + "loss": 1.1263, + "step": 2435 + }, + { + "epoch": 0.29291168159682557, + "grad_norm": 1.8575040235514337, + "learning_rate": 3.3178373272640394e-06, + "loss": 0.8641, + "step": 2436 + }, + { + "epoch": 0.2930319244874647, + "grad_norm": 2.6284214610062953, + "learning_rate": 3.3172512737910387e-06, + "loss": 1.0919, + "step": 2437 + }, + { + "epoch": 0.2931521673781038, + "grad_norm": 2.2703329695784995, + "learning_rate": 3.3166650204920674e-06, + "loss": 1.1231, + "step": 2438 + }, + { + "epoch": 0.29327241026874284, + "grad_norm": 1.6386470679132044, + "learning_rate": 3.316078567456059e-06, + "loss": 1.0536, + "step": 2439 + }, + { + "epoch": 0.29339265315938196, + "grad_norm": 1.8591518749068692, + "learning_rate": 3.3154919147719786e-06, + "loss": 0.9993, + "step": 2440 + }, + { + "epoch": 0.29351289605002107, + "grad_norm": 2.4391886412744235, + "learning_rate": 3.31490506252882e-06, + "loss": 1.1041, + "step": 2441 + }, + { + "epoch": 0.2936331389406601, + "grad_norm": 1.9090847762533656, + "learning_rate": 3.31431801081561e-06, + "loss": 1.084, + "step": 2442 + }, + { + "epoch": 0.29375338183129923, + "grad_norm": 0.9322755837074415, + "learning_rate": 3.313730759721402e-06, + "loss": 0.9181, + "step": 2443 + }, + { + "epoch": 0.29387362472193834, + "grad_norm": 2.0543339066070065, + "learning_rate": 3.313143309335282e-06, + "loss": 1.1044, + "step": 2444 + }, + { + "epoch": 0.2939938676125774, + "grad_norm": 2.70355318239314, + "learning_rate": 3.3125556597463665e-06, + "loss": 1.0853, + "step": 2445 + }, + { + "epoch": 0.2941141105032165, + "grad_norm": 1.609095576945207, + "learning_rate": 3.311967811043801e-06, + "loss": 0.9003, + "step": 2446 + }, + { + "epoch": 0.29423435339385556, + "grad_norm": 1.9452910517846846, + "learning_rate": 3.3113797633167617e-06, + "loss": 1.0529, + "step": 2447 + }, + { + "epoch": 0.2943545962844947, + "grad_norm": 2.189901396216868, + "learning_rate": 3.310791516654455e-06, + "loss": 0.9211, + "step": 2448 + }, + { + "epoch": 0.2944748391751338, + "grad_norm": 2.363072611528981, + "learning_rate": 3.3102030711461177e-06, + "loss": 1.0352, + "step": 2449 + }, + { + "epoch": 0.29459508206577284, + "grad_norm": 3.5710102626510993, + "learning_rate": 3.3096144268810156e-06, + "loss": 0.9178, + "step": 2450 + }, + { + "epoch": 0.29471532495641195, + "grad_norm": 2.521143198016, + "learning_rate": 3.3090255839484462e-06, + "loss": 0.9667, + "step": 2451 + }, + { + "epoch": 0.29483556784705106, + "grad_norm": 2.117286707957292, + "learning_rate": 3.3084365424377366e-06, + "loss": 1.0922, + "step": 2452 + }, + { + "epoch": 0.2949558107376901, + "grad_norm": 0.7353628291955197, + "learning_rate": 3.307847302438245e-06, + "loss": 0.7973, + "step": 2453 + }, + { + "epoch": 0.2950760536283292, + "grad_norm": 2.221107305216342, + "learning_rate": 3.3072578640393562e-06, + "loss": 1.0207, + "step": 2454 + }, + { + "epoch": 0.29519629651896834, + "grad_norm": 2.3782088035625226, + "learning_rate": 3.3066682273304886e-06, + "loss": 1.0363, + "step": 2455 + }, + { + "epoch": 0.2953165394096074, + "grad_norm": 2.16900280185376, + "learning_rate": 3.3060783924010904e-06, + "loss": 1.0272, + "step": 2456 + }, + { + "epoch": 0.2954367823002465, + "grad_norm": 2.0550198014319334, + "learning_rate": 3.3054883593406387e-06, + "loss": 1.0866, + "step": 2457 + }, + { + "epoch": 0.2955570251908856, + "grad_norm": 2.2389577538958765, + "learning_rate": 3.3048981282386404e-06, + "loss": 0.8906, + "step": 2458 + }, + { + "epoch": 0.29567726808152467, + "grad_norm": 2.85915119566846, + "learning_rate": 3.304307699184634e-06, + "loss": 1.0644, + "step": 2459 + }, + { + "epoch": 0.2957975109721638, + "grad_norm": 1.8280419224829678, + "learning_rate": 3.3037170722681866e-06, + "loss": 1.0252, + "step": 2460 + }, + { + "epoch": 0.29591775386280283, + "grad_norm": 1.9603084451909258, + "learning_rate": 3.3031262475788956e-06, + "loss": 0.9176, + "step": 2461 + }, + { + "epoch": 0.29603799675344195, + "grad_norm": 1.679859809021666, + "learning_rate": 3.3025352252063897e-06, + "loss": 0.9696, + "step": 2462 + }, + { + "epoch": 0.29615823964408106, + "grad_norm": 1.7771372916231283, + "learning_rate": 3.3019440052403252e-06, + "loss": 0.9859, + "step": 2463 + }, + { + "epoch": 0.2962784825347201, + "grad_norm": 1.981842527625752, + "learning_rate": 3.30135258777039e-06, + "loss": 0.9471, + "step": 2464 + }, + { + "epoch": 0.2963987254253592, + "grad_norm": 2.790113235144624, + "learning_rate": 3.3007609728863024e-06, + "loss": 0.9421, + "step": 2465 + }, + { + "epoch": 0.29651896831599833, + "grad_norm": 1.8732202348247091, + "learning_rate": 3.300169160677809e-06, + "loss": 0.9716, + "step": 2466 + }, + { + "epoch": 0.2966392112066374, + "grad_norm": 2.877904171502088, + "learning_rate": 3.2995771512346878e-06, + "loss": 1.0097, + "step": 2467 + }, + { + "epoch": 0.2967594540972765, + "grad_norm": 2.891332426368173, + "learning_rate": 3.298984944646746e-06, + "loss": 0.969, + "step": 2468 + }, + { + "epoch": 0.2968796969879156, + "grad_norm": 1.870950585863192, + "learning_rate": 3.298392541003822e-06, + "loss": 1.0524, + "step": 2469 + }, + { + "epoch": 0.29699993987855466, + "grad_norm": 1.7840646958316742, + "learning_rate": 3.2977999403957806e-06, + "loss": 1.1328, + "step": 2470 + }, + { + "epoch": 0.2971201827691938, + "grad_norm": 2.358712614942765, + "learning_rate": 3.2972071429125207e-06, + "loss": 0.909, + "step": 2471 + }, + { + "epoch": 0.2972404256598329, + "grad_norm": 2.176129462219638, + "learning_rate": 3.2966141486439682e-06, + "loss": 1.122, + "step": 2472 + }, + { + "epoch": 0.29736066855047194, + "grad_norm": 2.300356866268519, + "learning_rate": 3.29602095768008e-06, + "loss": 0.8821, + "step": 2473 + }, + { + "epoch": 0.29748091144111105, + "grad_norm": 1.9636770618279917, + "learning_rate": 3.2954275701108437e-06, + "loss": 0.8725, + "step": 2474 + }, + { + "epoch": 0.29760115433175016, + "grad_norm": 2.1224808065492424, + "learning_rate": 3.294833986026275e-06, + "loss": 0.9241, + "step": 2475 + }, + { + "epoch": 0.2977213972223892, + "grad_norm": 1.9261650646973565, + "learning_rate": 3.29424020551642e-06, + "loss": 1.0897, + "step": 2476 + }, + { + "epoch": 0.2978416401130283, + "grad_norm": 2.10698558968648, + "learning_rate": 3.2936462286713546e-06, + "loss": 0.9586, + "step": 2477 + }, + { + "epoch": 0.2979618830036674, + "grad_norm": 2.2192517640026352, + "learning_rate": 3.2930520555811846e-06, + "loss": 1.0112, + "step": 2478 + }, + { + "epoch": 0.2980821258943065, + "grad_norm": 1.9245483909067214, + "learning_rate": 3.292457686336046e-06, + "loss": 1.0364, + "step": 2479 + }, + { + "epoch": 0.2982023687849456, + "grad_norm": 0.8622364954699673, + "learning_rate": 3.291863121026105e-06, + "loss": 0.8504, + "step": 2480 + }, + { + "epoch": 0.29832261167558466, + "grad_norm": 2.4194672459976734, + "learning_rate": 3.2912683597415547e-06, + "loss": 1.0042, + "step": 2481 + }, + { + "epoch": 0.29844285456622377, + "grad_norm": 1.9851632542582307, + "learning_rate": 3.2906734025726213e-06, + "loss": 1.0165, + "step": 2482 + }, + { + "epoch": 0.2985630974568629, + "grad_norm": 1.9632709495313692, + "learning_rate": 3.290078249609559e-06, + "loss": 1.1153, + "step": 2483 + }, + { + "epoch": 0.29868334034750194, + "grad_norm": 2.079013032403551, + "learning_rate": 3.2894829009426514e-06, + "loss": 1.1183, + "step": 2484 + }, + { + "epoch": 0.29880358323814105, + "grad_norm": 2.144226274883696, + "learning_rate": 3.288887356662213e-06, + "loss": 1.0151, + "step": 2485 + }, + { + "epoch": 0.29892382612878016, + "grad_norm": 0.7763470725086802, + "learning_rate": 3.288291616858588e-06, + "loss": 0.8353, + "step": 2486 + }, + { + "epoch": 0.2990440690194192, + "grad_norm": 1.822365319569924, + "learning_rate": 3.287695681622149e-06, + "loss": 1.0076, + "step": 2487 + }, + { + "epoch": 0.2991643119100583, + "grad_norm": 3.8316600437004746, + "learning_rate": 3.2870995510432982e-06, + "loss": 1.0473, + "step": 2488 + }, + { + "epoch": 0.29928455480069743, + "grad_norm": 1.8767657548663115, + "learning_rate": 3.2865032252124697e-06, + "loss": 1.0093, + "step": 2489 + }, + { + "epoch": 0.2994047976913365, + "grad_norm": 1.6688065222836734, + "learning_rate": 3.2859067042201243e-06, + "loss": 1.0134, + "step": 2490 + }, + { + "epoch": 0.2995250405819756, + "grad_norm": 2.0599096967292523, + "learning_rate": 3.2853099881567544e-06, + "loss": 1.016, + "step": 2491 + }, + { + "epoch": 0.29964528347261465, + "grad_norm": 1.7550562972904702, + "learning_rate": 3.284713077112881e-06, + "loss": 1.0317, + "step": 2492 + }, + { + "epoch": 0.29976552636325376, + "grad_norm": 2.4785575652726157, + "learning_rate": 3.284115971179056e-06, + "loss": 1.1003, + "step": 2493 + }, + { + "epoch": 0.2998857692538929, + "grad_norm": 1.909620213722283, + "learning_rate": 3.283518670445859e-06, + "loss": 1.0221, + "step": 2494 + }, + { + "epoch": 0.30000601214453193, + "grad_norm": 0.7094950872604692, + "learning_rate": 3.2829211750038995e-06, + "loss": 0.7814, + "step": 2495 + }, + { + "epoch": 0.30012625503517104, + "grad_norm": 2.1465581268608096, + "learning_rate": 3.2823234849438183e-06, + "loss": 1.1282, + "step": 2496 + }, + { + "epoch": 0.30024649792581015, + "grad_norm": 3.556025863047071, + "learning_rate": 3.2817256003562836e-06, + "loss": 0.9989, + "step": 2497 + }, + { + "epoch": 0.3003667408164492, + "grad_norm": 2.1103280634103516, + "learning_rate": 3.281127521331995e-06, + "loss": 0.8985, + "step": 2498 + }, + { + "epoch": 0.3004869837070883, + "grad_norm": 0.8866913240871515, + "learning_rate": 3.2805292479616798e-06, + "loss": 0.8446, + "step": 2499 + }, + { + "epoch": 0.30060722659772743, + "grad_norm": 2.443430996339883, + "learning_rate": 3.2799307803360955e-06, + "loss": 1.1588, + "step": 2500 + }, + { + "epoch": 0.3007274694883665, + "grad_norm": 1.4794443254943894, + "learning_rate": 3.27933211854603e-06, + "loss": 1.0538, + "step": 2501 + }, + { + "epoch": 0.3008477123790056, + "grad_norm": 1.8309683573398623, + "learning_rate": 3.278733262682299e-06, + "loss": 1.1129, + "step": 2502 + }, + { + "epoch": 0.3009679552696447, + "grad_norm": 3.422113279642574, + "learning_rate": 3.2781342128357484e-06, + "loss": 1.0629, + "step": 2503 + }, + { + "epoch": 0.30108819816028376, + "grad_norm": 2.5205156635810986, + "learning_rate": 3.2775349690972547e-06, + "loss": 1.0442, + "step": 2504 + }, + { + "epoch": 0.30120844105092287, + "grad_norm": 0.7644157591563161, + "learning_rate": 3.276935531557722e-06, + "loss": 0.756, + "step": 2505 + }, + { + "epoch": 0.301328683941562, + "grad_norm": 2.5444229585479854, + "learning_rate": 3.2763359003080837e-06, + "loss": 1.0353, + "step": 2506 + }, + { + "epoch": 0.30144892683220104, + "grad_norm": 0.853291741868797, + "learning_rate": 3.2757360754393047e-06, + "loss": 0.863, + "step": 2507 + }, + { + "epoch": 0.30156916972284015, + "grad_norm": 2.429711541059969, + "learning_rate": 3.2751360570423767e-06, + "loss": 0.8858, + "step": 2508 + }, + { + "epoch": 0.3016894126134792, + "grad_norm": 3.33176319971602, + "learning_rate": 3.2745358452083236e-06, + "loss": 0.9988, + "step": 2509 + }, + { + "epoch": 0.3018096555041183, + "grad_norm": 1.4204577414736772, + "learning_rate": 3.2739354400281955e-06, + "loss": 1.0622, + "step": 2510 + }, + { + "epoch": 0.3019298983947574, + "grad_norm": 0.8687373908246188, + "learning_rate": 3.2733348415930744e-06, + "loss": 0.8748, + "step": 2511 + }, + { + "epoch": 0.3020501412853965, + "grad_norm": 1.9710642821127324, + "learning_rate": 3.27273404999407e-06, + "loss": 1.0427, + "step": 2512 + }, + { + "epoch": 0.3021703841760356, + "grad_norm": 0.9178819980616912, + "learning_rate": 3.272133065322322e-06, + "loss": 0.8425, + "step": 2513 + }, + { + "epoch": 0.3022906270666747, + "grad_norm": 1.8962417274450556, + "learning_rate": 3.271531887669e-06, + "loss": 1.0336, + "step": 2514 + }, + { + "epoch": 0.30241086995731375, + "grad_norm": 4.3964351344560395, + "learning_rate": 3.2709305171253015e-06, + "loss": 0.8732, + "step": 2515 + }, + { + "epoch": 0.30253111284795287, + "grad_norm": 1.8017207609672727, + "learning_rate": 3.2703289537824536e-06, + "loss": 1.0151, + "step": 2516 + }, + { + "epoch": 0.302651355738592, + "grad_norm": 2.5801199852596355, + "learning_rate": 3.269727197731714e-06, + "loss": 1.0234, + "step": 2517 + }, + { + "epoch": 0.30277159862923103, + "grad_norm": 1.8644668358011542, + "learning_rate": 3.269125249064367e-06, + "loss": 1.0151, + "step": 2518 + }, + { + "epoch": 0.30289184151987014, + "grad_norm": 1.678707599913967, + "learning_rate": 3.2685231078717297e-06, + "loss": 1.069, + "step": 2519 + }, + { + "epoch": 0.30301208441050925, + "grad_norm": 2.0256864208500915, + "learning_rate": 3.267920774245145e-06, + "loss": 0.9911, + "step": 2520 + }, + { + "epoch": 0.3031323273011483, + "grad_norm": 2.043064210021793, + "learning_rate": 3.2673182482759876e-06, + "loss": 1.0839, + "step": 2521 + }, + { + "epoch": 0.3032525701917874, + "grad_norm": 2.3360320107213313, + "learning_rate": 3.266715530055659e-06, + "loss": 0.8963, + "step": 2522 + }, + { + "epoch": 0.30337281308242653, + "grad_norm": 3.6073364842526288, + "learning_rate": 3.2661126196755927e-06, + "loss": 1.0406, + "step": 2523 + }, + { + "epoch": 0.3034930559730656, + "grad_norm": 0.9645308941015949, + "learning_rate": 3.265509517227248e-06, + "loss": 0.7972, + "step": 2524 + }, + { + "epoch": 0.3036132988637047, + "grad_norm": 1.8982399417766966, + "learning_rate": 3.264906222802115e-06, + "loss": 1.0487, + "step": 2525 + }, + { + "epoch": 0.30373354175434375, + "grad_norm": 2.153326845051032, + "learning_rate": 3.264302736491715e-06, + "loss": 1.0233, + "step": 2526 + }, + { + "epoch": 0.30385378464498286, + "grad_norm": 2.181592867336217, + "learning_rate": 3.263699058387594e-06, + "loss": 1.1107, + "step": 2527 + }, + { + "epoch": 0.30397402753562197, + "grad_norm": 3.262903420948287, + "learning_rate": 3.2630951885813315e-06, + "loss": 1.1472, + "step": 2528 + }, + { + "epoch": 0.304094270426261, + "grad_norm": 2.0378293806762415, + "learning_rate": 3.262491127164533e-06, + "loss": 1.02, + "step": 2529 + }, + { + "epoch": 0.30421451331690014, + "grad_norm": 2.216918941951718, + "learning_rate": 3.2618868742288337e-06, + "loss": 1.0374, + "step": 2530 + }, + { + "epoch": 0.30433475620753925, + "grad_norm": 1.9951494525199422, + "learning_rate": 3.261282429865899e-06, + "loss": 0.9616, + "step": 2531 + }, + { + "epoch": 0.3044549990981783, + "grad_norm": 1.629757958093282, + "learning_rate": 3.2606777941674225e-06, + "loss": 0.9622, + "step": 2532 + }, + { + "epoch": 0.3045752419888174, + "grad_norm": 2.2329643438134403, + "learning_rate": 3.2600729672251276e-06, + "loss": 1.0788, + "step": 2533 + }, + { + "epoch": 0.3046954848794565, + "grad_norm": 1.937914527110519, + "learning_rate": 3.259467949130765e-06, + "loss": 0.8901, + "step": 2534 + }, + { + "epoch": 0.3048157277700956, + "grad_norm": 3.367209025090712, + "learning_rate": 3.2588627399761164e-06, + "loss": 1.0665, + "step": 2535 + }, + { + "epoch": 0.3049359706607347, + "grad_norm": 1.8350069975134868, + "learning_rate": 3.2582573398529903e-06, + "loss": 0.946, + "step": 2536 + }, + { + "epoch": 0.3050562135513738, + "grad_norm": 2.5148967375895426, + "learning_rate": 3.2576517488532265e-06, + "loss": 0.9823, + "step": 2537 + }, + { + "epoch": 0.30517645644201286, + "grad_norm": 1.8206657099731196, + "learning_rate": 3.257045967068692e-06, + "loss": 1.1134, + "step": 2538 + }, + { + "epoch": 0.30529669933265197, + "grad_norm": 2.4286276906973563, + "learning_rate": 3.2564399945912848e-06, + "loss": 1.0612, + "step": 2539 + }, + { + "epoch": 0.305416942223291, + "grad_norm": 3.024471451162819, + "learning_rate": 3.2558338315129287e-06, + "loss": 1.0619, + "step": 2540 + }, + { + "epoch": 0.30553718511393013, + "grad_norm": 2.0329739810687784, + "learning_rate": 3.2552274779255785e-06, + "loss": 0.997, + "step": 2541 + }, + { + "epoch": 0.30565742800456924, + "grad_norm": 2.291227287017577, + "learning_rate": 3.2546209339212184e-06, + "loss": 1.0082, + "step": 2542 + }, + { + "epoch": 0.3057776708952083, + "grad_norm": 1.811502971687796, + "learning_rate": 3.25401419959186e-06, + "loss": 1.0161, + "step": 2543 + }, + { + "epoch": 0.3058979137858474, + "grad_norm": 2.0265883727175784, + "learning_rate": 3.253407275029545e-06, + "loss": 0.9999, + "step": 2544 + }, + { + "epoch": 0.3060181566764865, + "grad_norm": 1.8904481284898087, + "learning_rate": 3.2528001603263425e-06, + "loss": 1.0381, + "step": 2545 + }, + { + "epoch": 0.3061383995671256, + "grad_norm": 1.9381052733016801, + "learning_rate": 3.2521928555743514e-06, + "loss": 1.0537, + "step": 2546 + }, + { + "epoch": 0.3062586424577647, + "grad_norm": 2.0602448590129656, + "learning_rate": 3.2515853608657e-06, + "loss": 0.92, + "step": 2547 + }, + { + "epoch": 0.3063788853484038, + "grad_norm": 4.933088045637907, + "learning_rate": 3.250977676292545e-06, + "loss": 0.9896, + "step": 2548 + }, + { + "epoch": 0.30649912823904285, + "grad_norm": 2.1290865480561774, + "learning_rate": 3.2503698019470712e-06, + "loss": 1.034, + "step": 2549 + }, + { + "epoch": 0.30661937112968196, + "grad_norm": 2.565644730411915, + "learning_rate": 3.249761737921492e-06, + "loss": 1.025, + "step": 2550 + }, + { + "epoch": 0.30673961402032107, + "grad_norm": 2.184005234491971, + "learning_rate": 3.249153484308051e-06, + "loss": 0.9806, + "step": 2551 + }, + { + "epoch": 0.3068598569109601, + "grad_norm": 2.307712548118251, + "learning_rate": 3.2485450411990194e-06, + "loss": 1.0145, + "step": 2552 + }, + { + "epoch": 0.30698009980159924, + "grad_norm": 2.106357931660023, + "learning_rate": 3.2479364086866983e-06, + "loss": 1.0636, + "step": 2553 + }, + { + "epoch": 0.30710034269223835, + "grad_norm": 1.8712428278457502, + "learning_rate": 3.247327586863416e-06, + "loss": 1.0519, + "step": 2554 + }, + { + "epoch": 0.3072205855828774, + "grad_norm": 2.056644566487795, + "learning_rate": 3.2467185758215304e-06, + "loss": 1.0091, + "step": 2555 + }, + { + "epoch": 0.3073408284735165, + "grad_norm": 2.5953971504493856, + "learning_rate": 3.246109375653428e-06, + "loss": 1.0919, + "step": 2556 + }, + { + "epoch": 0.30746107136415557, + "grad_norm": 1.9486907060275571, + "learning_rate": 3.2454999864515243e-06, + "loss": 1.0236, + "step": 2557 + }, + { + "epoch": 0.3075813142547947, + "grad_norm": 1.9750296940563135, + "learning_rate": 3.244890408308263e-06, + "loss": 0.9265, + "step": 2558 + }, + { + "epoch": 0.3077015571454338, + "grad_norm": 2.0745082732187585, + "learning_rate": 3.2442806413161165e-06, + "loss": 0.8479, + "step": 2559 + }, + { + "epoch": 0.30782180003607285, + "grad_norm": 2.0375922152942274, + "learning_rate": 3.243670685567586e-06, + "loss": 0.9905, + "step": 2560 + }, + { + "epoch": 0.30794204292671196, + "grad_norm": 2.3407717283169958, + "learning_rate": 3.2430605411552012e-06, + "loss": 1.0403, + "step": 2561 + }, + { + "epoch": 0.30806228581735107, + "grad_norm": 0.9129362748647677, + "learning_rate": 3.2424502081715205e-06, + "loss": 0.9438, + "step": 2562 + }, + { + "epoch": 0.3081825287079901, + "grad_norm": 1.7168765868688916, + "learning_rate": 3.241839686709132e-06, + "loss": 1.0189, + "step": 2563 + }, + { + "epoch": 0.30830277159862923, + "grad_norm": 3.112727672415637, + "learning_rate": 3.2412289768606495e-06, + "loss": 1.0617, + "step": 2564 + }, + { + "epoch": 0.30842301448926834, + "grad_norm": 1.9589434816270184, + "learning_rate": 3.240618078718718e-06, + "loss": 1.0655, + "step": 2565 + }, + { + "epoch": 0.3085432573799074, + "grad_norm": 2.0142727833094143, + "learning_rate": 3.240006992376011e-06, + "loss": 0.9852, + "step": 2566 + }, + { + "epoch": 0.3086635002705465, + "grad_norm": 2.549332351795444, + "learning_rate": 3.2393957179252284e-06, + "loss": 1.0018, + "step": 2567 + }, + { + "epoch": 0.3087837431611856, + "grad_norm": 1.9558567931636457, + "learning_rate": 3.2387842554591016e-06, + "loss": 1.041, + "step": 2568 + }, + { + "epoch": 0.3089039860518247, + "grad_norm": 2.894182480200364, + "learning_rate": 3.238172605070388e-06, + "loss": 1.1172, + "step": 2569 + }, + { + "epoch": 0.3090242289424638, + "grad_norm": 2.3417049386730833, + "learning_rate": 3.2375607668518745e-06, + "loss": 1.0314, + "step": 2570 + }, + { + "epoch": 0.30914447183310284, + "grad_norm": 2.253582645238936, + "learning_rate": 3.236948740896377e-06, + "loss": 1.1402, + "step": 2571 + }, + { + "epoch": 0.30926471472374195, + "grad_norm": 1.4916559840553272, + "learning_rate": 3.2363365272967384e-06, + "loss": 1.0796, + "step": 2572 + }, + { + "epoch": 0.30938495761438106, + "grad_norm": 2.0854560210649127, + "learning_rate": 3.235724126145832e-06, + "loss": 1.0551, + "step": 2573 + }, + { + "epoch": 0.3095052005050201, + "grad_norm": 1.6990375290621873, + "learning_rate": 3.235111537536558e-06, + "loss": 1.0134, + "step": 2574 + }, + { + "epoch": 0.30962544339565923, + "grad_norm": 2.0265834316093274, + "learning_rate": 3.2344987615618456e-06, + "loss": 1.0681, + "step": 2575 + }, + { + "epoch": 0.30974568628629834, + "grad_norm": 1.6443853970542945, + "learning_rate": 3.2338857983146533e-06, + "loss": 1.0217, + "step": 2576 + }, + { + "epoch": 0.3098659291769374, + "grad_norm": 2.203895812768843, + "learning_rate": 3.233272647887966e-06, + "loss": 1.0012, + "step": 2577 + }, + { + "epoch": 0.3099861720675765, + "grad_norm": 2.2347675325478553, + "learning_rate": 3.2326593103747985e-06, + "loss": 1.1409, + "step": 2578 + }, + { + "epoch": 0.3101064149582156, + "grad_norm": 2.0447136336633953, + "learning_rate": 3.2320457858681936e-06, + "loss": 1.0907, + "step": 2579 + }, + { + "epoch": 0.31022665784885467, + "grad_norm": 2.3939124547747124, + "learning_rate": 3.2314320744612228e-06, + "loss": 1.0849, + "step": 2580 + }, + { + "epoch": 0.3103469007394938, + "grad_norm": 1.578640239944374, + "learning_rate": 3.2308181762469854e-06, + "loss": 1.0009, + "step": 2581 + }, + { + "epoch": 0.3104671436301329, + "grad_norm": 1.9376266192008693, + "learning_rate": 3.230204091318609e-06, + "loss": 1.0228, + "step": 2582 + }, + { + "epoch": 0.31058738652077195, + "grad_norm": 1.8889603024739587, + "learning_rate": 3.2295898197692503e-06, + "loss": 1.0825, + "step": 2583 + }, + { + "epoch": 0.31070762941141106, + "grad_norm": 1.7029547650012977, + "learning_rate": 3.228975361692094e-06, + "loss": 1.0296, + "step": 2584 + }, + { + "epoch": 0.31082787230205017, + "grad_norm": 2.2675165107235293, + "learning_rate": 3.228360717180352e-06, + "loss": 1.0358, + "step": 2585 + }, + { + "epoch": 0.3109481151926892, + "grad_norm": 0.8296971063722283, + "learning_rate": 3.227745886327266e-06, + "loss": 0.8316, + "step": 2586 + }, + { + "epoch": 0.31106835808332833, + "grad_norm": 0.8108778679886529, + "learning_rate": 3.227130869226105e-06, + "loss": 0.7962, + "step": 2587 + }, + { + "epoch": 0.3111886009739674, + "grad_norm": 2.382271126704593, + "learning_rate": 3.226515665970167e-06, + "loss": 1.0654, + "step": 2588 + }, + { + "epoch": 0.3113088438646065, + "grad_norm": 2.0823555241141207, + "learning_rate": 3.225900276652777e-06, + "loss": 1.0999, + "step": 2589 + }, + { + "epoch": 0.3114290867552456, + "grad_norm": 1.9437358549038797, + "learning_rate": 3.2252847013672906e-06, + "loss": 0.9963, + "step": 2590 + }, + { + "epoch": 0.31154932964588467, + "grad_norm": 2.059722070000162, + "learning_rate": 3.224668940207089e-06, + "loss": 0.9976, + "step": 2591 + }, + { + "epoch": 0.3116695725365238, + "grad_norm": 2.0048084867722484, + "learning_rate": 3.2240529932655828e-06, + "loss": 1.1042, + "step": 2592 + }, + { + "epoch": 0.3117898154271629, + "grad_norm": 3.6393985892829606, + "learning_rate": 3.223436860636211e-06, + "loss": 1.126, + "step": 2593 + }, + { + "epoch": 0.31191005831780194, + "grad_norm": 1.6403724112664304, + "learning_rate": 3.2228205424124403e-06, + "loss": 0.9824, + "step": 2594 + }, + { + "epoch": 0.31203030120844105, + "grad_norm": 2.74908397330106, + "learning_rate": 3.222204038687765e-06, + "loss": 0.9892, + "step": 2595 + }, + { + "epoch": 0.31215054409908016, + "grad_norm": 1.7145696955432417, + "learning_rate": 3.221587349555709e-06, + "loss": 1.1221, + "step": 2596 + }, + { + "epoch": 0.3122707869897192, + "grad_norm": 1.6538312993209947, + "learning_rate": 3.2209704751098236e-06, + "loss": 0.933, + "step": 2597 + }, + { + "epoch": 0.31239102988035833, + "grad_norm": 2.1387204796132178, + "learning_rate": 3.2203534154436875e-06, + "loss": 1.0615, + "step": 2598 + }, + { + "epoch": 0.31251127277099744, + "grad_norm": 2.0655201849263976, + "learning_rate": 3.2197361706509084e-06, + "loss": 0.9953, + "step": 2599 + }, + { + "epoch": 0.3126315156616365, + "grad_norm": 2.7877519878537385, + "learning_rate": 3.2191187408251228e-06, + "loss": 1.0734, + "step": 2600 + }, + { + "epoch": 0.3127517585522756, + "grad_norm": 2.2128167135852808, + "learning_rate": 3.218501126059993e-06, + "loss": 1.0264, + "step": 2601 + }, + { + "epoch": 0.31287200144291466, + "grad_norm": 2.0534555201072853, + "learning_rate": 3.2178833264492116e-06, + "loss": 1.0527, + "step": 2602 + }, + { + "epoch": 0.31299224433355377, + "grad_norm": 1.785459930833458, + "learning_rate": 3.217265342086498e-06, + "loss": 0.9979, + "step": 2603 + }, + { + "epoch": 0.3131124872241929, + "grad_norm": 2.640480443846293, + "learning_rate": 3.216647173065599e-06, + "loss": 0.9673, + "step": 2604 + }, + { + "epoch": 0.31323273011483194, + "grad_norm": 1.831598877466628, + "learning_rate": 3.216028819480292e-06, + "loss": 0.9737, + "step": 2605 + }, + { + "epoch": 0.31335297300547105, + "grad_norm": 3.931032712720229, + "learning_rate": 3.2154102814243793e-06, + "loss": 0.9915, + "step": 2606 + }, + { + "epoch": 0.31347321589611016, + "grad_norm": 2.124646325692195, + "learning_rate": 3.2147915589916937e-06, + "loss": 0.9059, + "step": 2607 + }, + { + "epoch": 0.3135934587867492, + "grad_norm": 2.2604098576268834, + "learning_rate": 3.2141726522760938e-06, + "loss": 1.0638, + "step": 2608 + }, + { + "epoch": 0.3137137016773883, + "grad_norm": 0.7112806560969241, + "learning_rate": 3.213553561371469e-06, + "loss": 0.7648, + "step": 2609 + }, + { + "epoch": 0.31383394456802743, + "grad_norm": 2.1950179475993363, + "learning_rate": 3.212934286371733e-06, + "loss": 1.1957, + "step": 2610 + }, + { + "epoch": 0.3139541874586665, + "grad_norm": 2.173586712890887, + "learning_rate": 3.2123148273708304e-06, + "loss": 1.0687, + "step": 2611 + }, + { + "epoch": 0.3140744303493056, + "grad_norm": 2.0098740499330394, + "learning_rate": 3.211695184462733e-06, + "loss": 1.0049, + "step": 2612 + }, + { + "epoch": 0.3141946732399447, + "grad_norm": 0.88909466086949, + "learning_rate": 3.2110753577414383e-06, + "loss": 0.8426, + "step": 2613 + }, + { + "epoch": 0.31431491613058377, + "grad_norm": 1.9906822352063531, + "learning_rate": 3.2104553473009757e-06, + "loss": 1.0282, + "step": 2614 + }, + { + "epoch": 0.3144351590212229, + "grad_norm": 2.250317657087615, + "learning_rate": 3.209835153235399e-06, + "loss": 0.9204, + "step": 2615 + }, + { + "epoch": 0.314555401911862, + "grad_norm": 1.906564780338501, + "learning_rate": 3.2092147756387916e-06, + "loss": 0.9145, + "step": 2616 + }, + { + "epoch": 0.31467564480250104, + "grad_norm": 2.1638201085483524, + "learning_rate": 3.208594214605264e-06, + "loss": 1.0718, + "step": 2617 + }, + { + "epoch": 0.31479588769314015, + "grad_norm": 2.4032833091866403, + "learning_rate": 3.2079734702289553e-06, + "loss": 1.0125, + "step": 2618 + }, + { + "epoch": 0.3149161305837792, + "grad_norm": 0.8061083868878801, + "learning_rate": 3.207352542604031e-06, + "loss": 0.8421, + "step": 2619 + }, + { + "epoch": 0.3150363734744183, + "grad_norm": 1.8117646665951963, + "learning_rate": 3.2067314318246864e-06, + "loss": 1.0222, + "step": 2620 + }, + { + "epoch": 0.31515661636505743, + "grad_norm": 1.7690347476002735, + "learning_rate": 3.206110137985143e-06, + "loss": 1.0132, + "step": 2621 + }, + { + "epoch": 0.3152768592556965, + "grad_norm": 2.280485573811433, + "learning_rate": 3.2054886611796505e-06, + "loss": 1.1624, + "step": 2622 + }, + { + "epoch": 0.3153971021463356, + "grad_norm": 0.8950534612256387, + "learning_rate": 3.204867001502487e-06, + "loss": 0.8733, + "step": 2623 + }, + { + "epoch": 0.3155173450369747, + "grad_norm": 2.589072180150318, + "learning_rate": 3.2042451590479567e-06, + "loss": 1.0435, + "step": 2624 + }, + { + "epoch": 0.31563758792761376, + "grad_norm": 3.747742545617618, + "learning_rate": 3.203623133910394e-06, + "loss": 1.1084, + "step": 2625 + }, + { + "epoch": 0.31575783081825287, + "grad_norm": 3.664836484584303, + "learning_rate": 3.203000926184158e-06, + "loss": 1.0122, + "step": 2626 + }, + { + "epoch": 0.315878073708892, + "grad_norm": 1.9700298690467928, + "learning_rate": 3.202378535963639e-06, + "loss": 1.0146, + "step": 2627 + }, + { + "epoch": 0.31599831659953104, + "grad_norm": 2.1311636898876873, + "learning_rate": 3.2017559633432516e-06, + "loss": 1.0781, + "step": 2628 + }, + { + "epoch": 0.31611855949017015, + "grad_norm": 1.9578746569631418, + "learning_rate": 3.2011332084174398e-06, + "loss": 0.902, + "step": 2629 + }, + { + "epoch": 0.31623880238080926, + "grad_norm": 1.6394046422712314, + "learning_rate": 3.2005102712806756e-06, + "loss": 1.1301, + "step": 2630 + }, + { + "epoch": 0.3163590452714483, + "grad_norm": 2.455389738981205, + "learning_rate": 3.1998871520274575e-06, + "loss": 0.967, + "step": 2631 + }, + { + "epoch": 0.3164792881620874, + "grad_norm": 2.038768410941608, + "learning_rate": 3.199263850752312e-06, + "loss": 1.0856, + "step": 2632 + }, + { + "epoch": 0.31659953105272653, + "grad_norm": 2.3277550729854655, + "learning_rate": 3.198640367549795e-06, + "loss": 1.0992, + "step": 2633 + }, + { + "epoch": 0.3167197739433656, + "grad_norm": 1.8285652999796516, + "learning_rate": 3.198016702514487e-06, + "loss": 1.0959, + "step": 2634 + }, + { + "epoch": 0.3168400168340047, + "grad_norm": 1.8282796680198106, + "learning_rate": 3.1973928557409972e-06, + "loss": 1.0863, + "step": 2635 + }, + { + "epoch": 0.31696025972464376, + "grad_norm": 2.235109328597801, + "learning_rate": 3.1967688273239636e-06, + "loss": 0.9484, + "step": 2636 + }, + { + "epoch": 0.31708050261528287, + "grad_norm": 2.1567696207855755, + "learning_rate": 3.1961446173580503e-06, + "loss": 1.0544, + "step": 2637 + }, + { + "epoch": 0.317200745505922, + "grad_norm": 1.7328041454832694, + "learning_rate": 3.1955202259379502e-06, + "loss": 1.0112, + "step": 2638 + }, + { + "epoch": 0.31732098839656103, + "grad_norm": 1.9081856327563147, + "learning_rate": 3.194895653158381e-06, + "loss": 1.068, + "step": 2639 + }, + { + "epoch": 0.31744123128720014, + "grad_norm": 0.7685023754110463, + "learning_rate": 3.194270899114093e-06, + "loss": 0.7928, + "step": 2640 + }, + { + "epoch": 0.31756147417783925, + "grad_norm": 1.8659027659604996, + "learning_rate": 3.193645963899858e-06, + "loss": 1.0574, + "step": 2641 + }, + { + "epoch": 0.3176817170684783, + "grad_norm": 2.0820219744677884, + "learning_rate": 3.193020847610479e-06, + "loss": 1.0786, + "step": 2642 + }, + { + "epoch": 0.3178019599591174, + "grad_norm": 2.4986618275278323, + "learning_rate": 3.192395550340787e-06, + "loss": 0.9492, + "step": 2643 + }, + { + "epoch": 0.31792220284975653, + "grad_norm": 2.078649583732059, + "learning_rate": 3.191770072185638e-06, + "loss": 1.0042, + "step": 2644 + }, + { + "epoch": 0.3180424457403956, + "grad_norm": 2.4738959267759335, + "learning_rate": 3.191144413239916e-06, + "loss": 0.9607, + "step": 2645 + }, + { + "epoch": 0.3181626886310347, + "grad_norm": 3.298396911805063, + "learning_rate": 3.190518573598534e-06, + "loss": 1.1162, + "step": 2646 + }, + { + "epoch": 0.3182829315216738, + "grad_norm": 1.7073546917190232, + "learning_rate": 3.1898925533564308e-06, + "loss": 1.0108, + "step": 2647 + }, + { + "epoch": 0.31840317441231286, + "grad_norm": 2.122398129951467, + "learning_rate": 3.1892663526085733e-06, + "loss": 0.8761, + "step": 2648 + }, + { + "epoch": 0.31852341730295197, + "grad_norm": 0.7616839278649067, + "learning_rate": 3.188639971449956e-06, + "loss": 0.8073, + "step": 2649 + }, + { + "epoch": 0.318643660193591, + "grad_norm": 2.004238881841034, + "learning_rate": 3.1880134099756e-06, + "loss": 0.9628, + "step": 2650 + }, + { + "epoch": 0.31876390308423014, + "grad_norm": 1.8906214690372503, + "learning_rate": 3.1873866682805535e-06, + "loss": 0.9313, + "step": 2651 + }, + { + "epoch": 0.31888414597486925, + "grad_norm": 1.8160547059847605, + "learning_rate": 3.186759746459894e-06, + "loss": 1.1221, + "step": 2652 + }, + { + "epoch": 0.3190043888655083, + "grad_norm": 1.9675298876110061, + "learning_rate": 3.1861326446087246e-06, + "loss": 1.032, + "step": 2653 + }, + { + "epoch": 0.3191246317561474, + "grad_norm": 2.115763843662822, + "learning_rate": 3.1855053628221763e-06, + "loss": 0.956, + "step": 2654 + }, + { + "epoch": 0.3192448746467865, + "grad_norm": 2.526457028142626, + "learning_rate": 3.184877901195407e-06, + "loss": 1.1391, + "step": 2655 + }, + { + "epoch": 0.3193651175374256, + "grad_norm": 0.7994051763402744, + "learning_rate": 3.184250259823602e-06, + "loss": 0.8661, + "step": 2656 + }, + { + "epoch": 0.3194853604280647, + "grad_norm": 2.10184678207635, + "learning_rate": 3.183622438801974e-06, + "loss": 1.0519, + "step": 2657 + }, + { + "epoch": 0.3196056033187038, + "grad_norm": 2.027862306107527, + "learning_rate": 3.1829944382257637e-06, + "loss": 0.9877, + "step": 2658 + }, + { + "epoch": 0.31972584620934286, + "grad_norm": 2.2464696951001626, + "learning_rate": 3.1823662581902373e-06, + "loss": 1.0569, + "step": 2659 + }, + { + "epoch": 0.31984608909998197, + "grad_norm": 2.75837161526425, + "learning_rate": 3.1817378987906896e-06, + "loss": 0.9848, + "step": 2660 + }, + { + "epoch": 0.3199663319906211, + "grad_norm": 1.9578504846393754, + "learning_rate": 3.181109360122442e-06, + "loss": 1.0353, + "step": 2661 + }, + { + "epoch": 0.32008657488126013, + "grad_norm": 2.041573450983755, + "learning_rate": 3.1804806422808445e-06, + "loss": 1.0181, + "step": 2662 + }, + { + "epoch": 0.32020681777189924, + "grad_norm": 1.7328598002246163, + "learning_rate": 3.1798517453612714e-06, + "loss": 0.9716, + "step": 2663 + }, + { + "epoch": 0.32032706066253835, + "grad_norm": 2.0313228887538397, + "learning_rate": 3.1792226694591265e-06, + "loss": 0.9906, + "step": 2664 + }, + { + "epoch": 0.3204473035531774, + "grad_norm": 1.9958358568190175, + "learning_rate": 3.178593414669841e-06, + "loss": 1.0413, + "step": 2665 + }, + { + "epoch": 0.3205675464438165, + "grad_norm": 2.105079968852122, + "learning_rate": 3.1779639810888707e-06, + "loss": 0.9395, + "step": 2666 + }, + { + "epoch": 0.3206877893344556, + "grad_norm": 2.4395667875029075, + "learning_rate": 3.1773343688117013e-06, + "loss": 0.9988, + "step": 2667 + }, + { + "epoch": 0.3208080322250947, + "grad_norm": 2.695999456020719, + "learning_rate": 3.1767045779338445e-06, + "loss": 1.0769, + "step": 2668 + }, + { + "epoch": 0.3209282751157338, + "grad_norm": 2.666407473206484, + "learning_rate": 3.176074608550839e-06, + "loss": 1.152, + "step": 2669 + }, + { + "epoch": 0.32104851800637285, + "grad_norm": 2.4800564641062164, + "learning_rate": 3.17544446075825e-06, + "loss": 1.0592, + "step": 2670 + }, + { + "epoch": 0.32116876089701196, + "grad_norm": 1.663435768701289, + "learning_rate": 3.174814134651671e-06, + "loss": 0.9488, + "step": 2671 + }, + { + "epoch": 0.3212890037876511, + "grad_norm": 3.3970527721727066, + "learning_rate": 3.1741836303267215e-06, + "loss": 1.0437, + "step": 2672 + }, + { + "epoch": 0.32140924667829013, + "grad_norm": 2.5708158516147783, + "learning_rate": 3.1735529478790496e-06, + "loss": 0.9888, + "step": 2673 + }, + { + "epoch": 0.32152948956892924, + "grad_norm": 1.9130480510860568, + "learning_rate": 3.172922087404328e-06, + "loss": 1.0321, + "step": 2674 + }, + { + "epoch": 0.32164973245956835, + "grad_norm": 0.7846875286216229, + "learning_rate": 3.1722910489982586e-06, + "loss": 0.7906, + "step": 2675 + }, + { + "epoch": 0.3217699753502074, + "grad_norm": 1.7024752576153224, + "learning_rate": 3.1716598327565694e-06, + "loss": 1.0398, + "step": 2676 + }, + { + "epoch": 0.3218902182408465, + "grad_norm": 1.5866155067712229, + "learning_rate": 3.171028438775015e-06, + "loss": 1.0786, + "step": 2677 + }, + { + "epoch": 0.3220104611314856, + "grad_norm": 2.1291920321372606, + "learning_rate": 3.170396867149377e-06, + "loss": 1.0788, + "step": 2678 + }, + { + "epoch": 0.3221307040221247, + "grad_norm": 1.8265747931387342, + "learning_rate": 3.1697651179754653e-06, + "loss": 1.1035, + "step": 2679 + }, + { + "epoch": 0.3222509469127638, + "grad_norm": 1.9152178955614, + "learning_rate": 3.1691331913491153e-06, + "loss": 0.9663, + "step": 2680 + }, + { + "epoch": 0.32237118980340285, + "grad_norm": 2.066946150227299, + "learning_rate": 3.1685010873661898e-06, + "loss": 1.084, + "step": 2681 + }, + { + "epoch": 0.32249143269404196, + "grad_norm": 1.8867464873050084, + "learning_rate": 3.167868806122578e-06, + "loss": 1.0354, + "step": 2682 + }, + { + "epoch": 0.32261167558468107, + "grad_norm": 1.950475375601142, + "learning_rate": 3.1672363477141968e-06, + "loss": 0.8988, + "step": 2683 + }, + { + "epoch": 0.3227319184753201, + "grad_norm": 3.273337132774784, + "learning_rate": 3.1666037122369903e-06, + "loss": 1.085, + "step": 2684 + }, + { + "epoch": 0.32285216136595923, + "grad_norm": 2.3002083393773303, + "learning_rate": 3.165970899786928e-06, + "loss": 1.1018, + "step": 2685 + }, + { + "epoch": 0.32297240425659834, + "grad_norm": 1.740517675628376, + "learning_rate": 3.1653379104600067e-06, + "loss": 0.9889, + "step": 2686 + }, + { + "epoch": 0.3230926471472374, + "grad_norm": 1.6040655162016713, + "learning_rate": 3.164704744352251e-06, + "loss": 0.9329, + "step": 2687 + }, + { + "epoch": 0.3232128900378765, + "grad_norm": 1.8822343500555536, + "learning_rate": 3.164071401559713e-06, + "loss": 1.0484, + "step": 2688 + }, + { + "epoch": 0.3233331329285156, + "grad_norm": 1.7792324217795914, + "learning_rate": 3.1634378821784674e-06, + "loss": 0.9496, + "step": 2689 + }, + { + "epoch": 0.3234533758191547, + "grad_norm": 2.4743869985668687, + "learning_rate": 3.1628041863046208e-06, + "loss": 0.9781, + "step": 2690 + }, + { + "epoch": 0.3235736187097938, + "grad_norm": 2.119949959758169, + "learning_rate": 3.162170314034304e-06, + "loss": 1.1458, + "step": 2691 + }, + { + "epoch": 0.3236938616004329, + "grad_norm": 1.6172647503816895, + "learning_rate": 3.1615362654636738e-06, + "loss": 1.0485, + "step": 2692 + }, + { + "epoch": 0.32381410449107195, + "grad_norm": 1.9629344726523608, + "learning_rate": 3.1609020406889163e-06, + "loss": 1.1094, + "step": 2693 + }, + { + "epoch": 0.32393434738171106, + "grad_norm": 1.7090723331159474, + "learning_rate": 3.1602676398062416e-06, + "loss": 1.0866, + "step": 2694 + }, + { + "epoch": 0.3240545902723502, + "grad_norm": 1.8856296594629827, + "learning_rate": 3.1596330629118886e-06, + "loss": 0.8554, + "step": 2695 + }, + { + "epoch": 0.32417483316298923, + "grad_norm": 2.1707328153196035, + "learning_rate": 3.1589983101021223e-06, + "loss": 0.9703, + "step": 2696 + }, + { + "epoch": 0.32429507605362834, + "grad_norm": 2.434985698054805, + "learning_rate": 3.1583633814732337e-06, + "loss": 1.0868, + "step": 2697 + }, + { + "epoch": 0.3244153189442674, + "grad_norm": 2.5315807738643477, + "learning_rate": 3.157728277121541e-06, + "loss": 0.9611, + "step": 2698 + }, + { + "epoch": 0.3245355618349065, + "grad_norm": 2.6550639477803504, + "learning_rate": 3.1570929971433897e-06, + "loss": 1.02, + "step": 2699 + }, + { + "epoch": 0.3246558047255456, + "grad_norm": 2.0240743805709056, + "learning_rate": 3.1564575416351504e-06, + "loss": 1.0734, + "step": 2700 + }, + { + "epoch": 0.32477604761618467, + "grad_norm": 2.205215111963674, + "learning_rate": 3.155821910693221e-06, + "loss": 0.9786, + "step": 2701 + }, + { + "epoch": 0.3248962905068238, + "grad_norm": 1.6166363205306524, + "learning_rate": 3.1551861044140275e-06, + "loss": 1.0999, + "step": 2702 + }, + { + "epoch": 0.3250165333974629, + "grad_norm": 1.7320786128539192, + "learning_rate": 3.15455012289402e-06, + "loss": 1.0106, + "step": 2703 + }, + { + "epoch": 0.32513677628810195, + "grad_norm": 1.7947936899532073, + "learning_rate": 3.153913966229677e-06, + "loss": 1.0786, + "step": 2704 + }, + { + "epoch": 0.32525701917874106, + "grad_norm": 0.6562516348682202, + "learning_rate": 3.1532776345175027e-06, + "loss": 0.7407, + "step": 2705 + }, + { + "epoch": 0.32537726206938017, + "grad_norm": 1.866319015463452, + "learning_rate": 3.1526411278540285e-06, + "loss": 1.0281, + "step": 2706 + }, + { + "epoch": 0.3254975049600192, + "grad_norm": 2.520176531345673, + "learning_rate": 3.1520044463358116e-06, + "loss": 1.0536, + "step": 2707 + }, + { + "epoch": 0.32561774785065833, + "grad_norm": 1.6619540185638582, + "learning_rate": 3.151367590059436e-06, + "loss": 1.0419, + "step": 2708 + }, + { + "epoch": 0.32573799074129745, + "grad_norm": 2.081956701035488, + "learning_rate": 3.1507305591215117e-06, + "loss": 1.1059, + "step": 2709 + }, + { + "epoch": 0.3258582336319365, + "grad_norm": 0.6766997473441592, + "learning_rate": 3.150093353618677e-06, + "loss": 0.7953, + "step": 2710 + }, + { + "epoch": 0.3259784765225756, + "grad_norm": 2.610748386439908, + "learning_rate": 3.149455973647596e-06, + "loss": 1.1299, + "step": 2711 + }, + { + "epoch": 0.32609871941321467, + "grad_norm": 3.9298969311581655, + "learning_rate": 3.1488184193049563e-06, + "loss": 1.014, + "step": 2712 + }, + { + "epoch": 0.3262189623038538, + "grad_norm": 1.5776050579798555, + "learning_rate": 3.1481806906874767e-06, + "loss": 0.9605, + "step": 2713 + }, + { + "epoch": 0.3263392051944929, + "grad_norm": 1.6238346322713613, + "learning_rate": 3.147542787891899e-06, + "loss": 1.1155, + "step": 2714 + }, + { + "epoch": 0.32645944808513194, + "grad_norm": 2.7166534758951615, + "learning_rate": 3.1469047110149926e-06, + "loss": 0.9891, + "step": 2715 + }, + { + "epoch": 0.32657969097577105, + "grad_norm": 2.1285402315803323, + "learning_rate": 3.146266460153554e-06, + "loss": 1.0903, + "step": 2716 + }, + { + "epoch": 0.32669993386641016, + "grad_norm": 1.7392881070582435, + "learning_rate": 3.145628035404404e-06, + "loss": 1.0429, + "step": 2717 + }, + { + "epoch": 0.3268201767570492, + "grad_norm": 0.8997129923106381, + "learning_rate": 3.1449894368643922e-06, + "loss": 0.8123, + "step": 2718 + }, + { + "epoch": 0.32694041964768833, + "grad_norm": 1.5249072687690641, + "learning_rate": 3.1443506646303934e-06, + "loss": 0.9532, + "step": 2719 + }, + { + "epoch": 0.32706066253832744, + "grad_norm": 2.565620383329338, + "learning_rate": 3.1437117187993086e-06, + "loss": 0.9146, + "step": 2720 + }, + { + "epoch": 0.3271809054289665, + "grad_norm": 1.6142170480333866, + "learning_rate": 3.143072599468065e-06, + "loss": 1.0392, + "step": 2721 + }, + { + "epoch": 0.3273011483196056, + "grad_norm": 1.6077713615411307, + "learning_rate": 3.1424333067336174e-06, + "loss": 0.9948, + "step": 2722 + }, + { + "epoch": 0.3274213912102447, + "grad_norm": 1.7946959177942685, + "learning_rate": 3.141793840692945e-06, + "loss": 1.0214, + "step": 2723 + }, + { + "epoch": 0.32754163410088377, + "grad_norm": 2.637984179850609, + "learning_rate": 3.1411542014430553e-06, + "loss": 0.8517, + "step": 2724 + }, + { + "epoch": 0.3276618769915229, + "grad_norm": 2.3940072660917506, + "learning_rate": 3.1405143890809804e-06, + "loss": 1.0567, + "step": 2725 + }, + { + "epoch": 0.327782119882162, + "grad_norm": 2.42915304015467, + "learning_rate": 3.1398744037037796e-06, + "loss": 0.9357, + "step": 2726 + }, + { + "epoch": 0.32790236277280105, + "grad_norm": 1.9326331632858826, + "learning_rate": 3.139234245408538e-06, + "loss": 1.0758, + "step": 2727 + }, + { + "epoch": 0.32802260566344016, + "grad_norm": 1.3381730053133463, + "learning_rate": 3.1385939142923666e-06, + "loss": 0.998, + "step": 2728 + }, + { + "epoch": 0.3281428485540792, + "grad_norm": 2.14398708978723, + "learning_rate": 3.137953410452405e-06, + "loss": 1.0194, + "step": 2729 + }, + { + "epoch": 0.3282630914447183, + "grad_norm": 1.8365213318086209, + "learning_rate": 3.1373127339858146e-06, + "loss": 0.982, + "step": 2730 + }, + { + "epoch": 0.32838333433535744, + "grad_norm": 1.97448090117417, + "learning_rate": 3.136671884989787e-06, + "loss": 0.9869, + "step": 2731 + }, + { + "epoch": 0.3285035772259965, + "grad_norm": 2.5481594100024787, + "learning_rate": 3.1360308635615383e-06, + "loss": 1.1047, + "step": 2732 + }, + { + "epoch": 0.3286238201166356, + "grad_norm": 1.9543818587819304, + "learning_rate": 3.135389669798311e-06, + "loss": 1.0276, + "step": 2733 + }, + { + "epoch": 0.3287440630072747, + "grad_norm": 1.8318422785335127, + "learning_rate": 3.134748303797373e-06, + "loss": 1.0372, + "step": 2734 + }, + { + "epoch": 0.32886430589791377, + "grad_norm": 3.3541146189440454, + "learning_rate": 3.1341067656560203e-06, + "loss": 1.0467, + "step": 2735 + }, + { + "epoch": 0.3289845487885529, + "grad_norm": 2.1421454383065175, + "learning_rate": 3.133465055471572e-06, + "loss": 1.1016, + "step": 2736 + }, + { + "epoch": 0.329104791679192, + "grad_norm": 2.7659617935542964, + "learning_rate": 3.1328231733413767e-06, + "loss": 0.8956, + "step": 2737 + }, + { + "epoch": 0.32922503456983104, + "grad_norm": 2.363743861294849, + "learning_rate": 3.1321811193628067e-06, + "loss": 1.1496, + "step": 2738 + }, + { + "epoch": 0.32934527746047015, + "grad_norm": 2.1146440043147052, + "learning_rate": 3.131538893633261e-06, + "loss": 0.9354, + "step": 2739 + }, + { + "epoch": 0.32946552035110926, + "grad_norm": 1.9438921174227923, + "learning_rate": 3.130896496250165e-06, + "loss": 1.0142, + "step": 2740 + }, + { + "epoch": 0.3295857632417483, + "grad_norm": 2.1491507161339753, + "learning_rate": 3.1302539273109693e-06, + "loss": 1.1022, + "step": 2741 + }, + { + "epoch": 0.32970600613238743, + "grad_norm": 1.870897254899668, + "learning_rate": 3.1296111869131513e-06, + "loss": 1.0414, + "step": 2742 + }, + { + "epoch": 0.32982624902302654, + "grad_norm": 1.9930614873049814, + "learning_rate": 3.1289682751542153e-06, + "loss": 1.0931, + "step": 2743 + }, + { + "epoch": 0.3299464919136656, + "grad_norm": 2.023316957867306, + "learning_rate": 3.1283251921316883e-06, + "loss": 0.9527, + "step": 2744 + }, + { + "epoch": 0.3300667348043047, + "grad_norm": 2.270639382108653, + "learning_rate": 3.1276819379431277e-06, + "loss": 1.0474, + "step": 2745 + }, + { + "epoch": 0.33018697769494376, + "grad_norm": 1.9374992616713562, + "learning_rate": 3.1270385126861134e-06, + "loss": 0.9919, + "step": 2746 + }, + { + "epoch": 0.3303072205855829, + "grad_norm": 1.8544167839256764, + "learning_rate": 3.1263949164582533e-06, + "loss": 1.0594, + "step": 2747 + }, + { + "epoch": 0.330427463476222, + "grad_norm": 4.555202971876252, + "learning_rate": 3.1257511493571797e-06, + "loss": 1.0154, + "step": 2748 + }, + { + "epoch": 0.33054770636686104, + "grad_norm": 1.8246622138762902, + "learning_rate": 3.125107211480552e-06, + "loss": 1.0236, + "step": 2749 + }, + { + "epoch": 0.33066794925750015, + "grad_norm": 2.071345463856681, + "learning_rate": 3.124463102926054e-06, + "loss": 1.0359, + "step": 2750 + }, + { + "epoch": 0.33078819214813926, + "grad_norm": 0.7715547167422908, + "learning_rate": 3.1238188237913984e-06, + "loss": 0.8544, + "step": 2751 + }, + { + "epoch": 0.3309084350387783, + "grad_norm": 2.429982256412481, + "learning_rate": 3.1231743741743202e-06, + "loss": 1.001, + "step": 2752 + }, + { + "epoch": 0.3310286779294174, + "grad_norm": 2.56316757810473, + "learning_rate": 3.122529754172582e-06, + "loss": 1.0719, + "step": 2753 + }, + { + "epoch": 0.33114892082005654, + "grad_norm": 3.466637182721667, + "learning_rate": 3.1218849638839736e-06, + "loss": 0.9651, + "step": 2754 + }, + { + "epoch": 0.3312691637106956, + "grad_norm": 1.9175040309420992, + "learning_rate": 3.121240003406307e-06, + "loss": 1.0208, + "step": 2755 + }, + { + "epoch": 0.3313894066013347, + "grad_norm": 2.178598111188023, + "learning_rate": 3.120594872837425e-06, + "loss": 0.9649, + "step": 2756 + }, + { + "epoch": 0.3315096494919738, + "grad_norm": 0.8356846935147206, + "learning_rate": 3.1199495722751906e-06, + "loss": 0.8611, + "step": 2757 + }, + { + "epoch": 0.33162989238261287, + "grad_norm": 1.785949263780086, + "learning_rate": 3.1193041018174972e-06, + "loss": 1.0752, + "step": 2758 + }, + { + "epoch": 0.331750135273252, + "grad_norm": 2.0529373896101117, + "learning_rate": 3.118658461562261e-06, + "loss": 1.187, + "step": 2759 + }, + { + "epoch": 0.33187037816389103, + "grad_norm": 1.6753687495384209, + "learning_rate": 3.118012651607426e-06, + "loss": 1.086, + "step": 2760 + }, + { + "epoch": 0.33199062105453014, + "grad_norm": 2.735932435227635, + "learning_rate": 3.1173666720509603e-06, + "loss": 1.0728, + "step": 2761 + }, + { + "epoch": 0.33211086394516925, + "grad_norm": 2.1364054057835857, + "learning_rate": 3.116720522990859e-06, + "loss": 0.9188, + "step": 2762 + }, + { + "epoch": 0.3322311068358083, + "grad_norm": 1.9740051492123372, + "learning_rate": 3.116074204525142e-06, + "loss": 0.8587, + "step": 2763 + }, + { + "epoch": 0.3323513497264474, + "grad_norm": 1.9994077401612274, + "learning_rate": 3.1154277167518553e-06, + "loss": 1.0751, + "step": 2764 + }, + { + "epoch": 0.33247159261708653, + "grad_norm": 0.780390075445101, + "learning_rate": 3.114781059769072e-06, + "loss": 0.8325, + "step": 2765 + }, + { + "epoch": 0.3325918355077256, + "grad_norm": 2.650657331165604, + "learning_rate": 3.1141342336748874e-06, + "loss": 0.9133, + "step": 2766 + }, + { + "epoch": 0.3327120783983647, + "grad_norm": 1.545534574252139, + "learning_rate": 3.1134872385674253e-06, + "loss": 1.0598, + "step": 2767 + }, + { + "epoch": 0.3328323212890038, + "grad_norm": 1.797444195365261, + "learning_rate": 3.1128400745448353e-06, + "loss": 1.0936, + "step": 2768 + }, + { + "epoch": 0.33295256417964286, + "grad_norm": 2.3129806019142363, + "learning_rate": 3.11219274170529e-06, + "loss": 0.8735, + "step": 2769 + }, + { + "epoch": 0.333072807070282, + "grad_norm": 1.7962916215017926, + "learning_rate": 3.1115452401469903e-06, + "loss": 1.0515, + "step": 2770 + }, + { + "epoch": 0.3331930499609211, + "grad_norm": 2.503551916322765, + "learning_rate": 3.1108975699681613e-06, + "loss": 1.1042, + "step": 2771 + }, + { + "epoch": 0.33331329285156014, + "grad_norm": 1.917557371193902, + "learning_rate": 3.1102497312670542e-06, + "loss": 0.9555, + "step": 2772 + }, + { + "epoch": 0.33343353574219925, + "grad_norm": 1.979719573305326, + "learning_rate": 3.109601724141946e-06, + "loss": 1.0419, + "step": 2773 + }, + { + "epoch": 0.33355377863283836, + "grad_norm": 1.7883151963171324, + "learning_rate": 3.108953548691138e-06, + "loss": 0.9198, + "step": 2774 + }, + { + "epoch": 0.3336740215234774, + "grad_norm": 2.283958916924295, + "learning_rate": 3.108305205012959e-06, + "loss": 0.9608, + "step": 2775 + }, + { + "epoch": 0.3337942644141165, + "grad_norm": 2.0582924899271666, + "learning_rate": 3.107656693205761e-06, + "loss": 1.1145, + "step": 2776 + }, + { + "epoch": 0.3339145073047556, + "grad_norm": 3.29059686458547, + "learning_rate": 3.107008013367924e-06, + "loss": 0.9426, + "step": 2777 + }, + { + "epoch": 0.3340347501953947, + "grad_norm": 6.2545152757210944, + "learning_rate": 3.1063591655978507e-06, + "loss": 1.1046, + "step": 2778 + }, + { + "epoch": 0.3341549930860338, + "grad_norm": 1.9199853306448353, + "learning_rate": 3.105710149993972e-06, + "loss": 1.0338, + "step": 2779 + }, + { + "epoch": 0.33427523597667286, + "grad_norm": 1.8096753175905438, + "learning_rate": 3.1050609666547427e-06, + "loss": 1.0894, + "step": 2780 + }, + { + "epoch": 0.33439547886731197, + "grad_norm": 2.12235645343467, + "learning_rate": 3.104411615678644e-06, + "loss": 1.0095, + "step": 2781 + }, + { + "epoch": 0.3345157217579511, + "grad_norm": 4.105930988339412, + "learning_rate": 3.1037620971641803e-06, + "loss": 0.9732, + "step": 2782 + }, + { + "epoch": 0.33463596464859013, + "grad_norm": 3.8121018749045623, + "learning_rate": 3.1031124112098844e-06, + "loss": 0.8854, + "step": 2783 + }, + { + "epoch": 0.33475620753922924, + "grad_norm": 1.9599265117881608, + "learning_rate": 3.1024625579143127e-06, + "loss": 0.9604, + "step": 2784 + }, + { + "epoch": 0.33487645042986836, + "grad_norm": 2.092920053448811, + "learning_rate": 3.101812537376048e-06, + "loss": 0.9656, + "step": 2785 + }, + { + "epoch": 0.3349966933205074, + "grad_norm": 2.1439775262765024, + "learning_rate": 3.1011623496936973e-06, + "loss": 1.0832, + "step": 2786 + }, + { + "epoch": 0.3351169362111465, + "grad_norm": 1.8452828710956841, + "learning_rate": 3.100511994965893e-06, + "loss": 0.9285, + "step": 2787 + }, + { + "epoch": 0.33523717910178563, + "grad_norm": 1.8231238401851948, + "learning_rate": 3.0998614732912947e-06, + "loss": 1.0807, + "step": 2788 + }, + { + "epoch": 0.3353574219924247, + "grad_norm": 1.9187287419536339, + "learning_rate": 3.0992107847685855e-06, + "loss": 0.9177, + "step": 2789 + }, + { + "epoch": 0.3354776648830638, + "grad_norm": 2.185163503504564, + "learning_rate": 3.0985599294964736e-06, + "loss": 1.031, + "step": 2790 + }, + { + "epoch": 0.33559790777370285, + "grad_norm": 2.065902446633516, + "learning_rate": 3.097908907573695e-06, + "loss": 0.9333, + "step": 2791 + }, + { + "epoch": 0.33571815066434196, + "grad_norm": 1.9738205297140132, + "learning_rate": 3.0972577190990067e-06, + "loss": 1.1315, + "step": 2792 + }, + { + "epoch": 0.3358383935549811, + "grad_norm": 2.093272140402241, + "learning_rate": 3.096606364171196e-06, + "loss": 1.0394, + "step": 2793 + }, + { + "epoch": 0.33595863644562013, + "grad_norm": 1.989622252185168, + "learning_rate": 3.0959548428890703e-06, + "loss": 1.0922, + "step": 2794 + }, + { + "epoch": 0.33607887933625924, + "grad_norm": 1.70642316449552, + "learning_rate": 3.095303155351468e-06, + "loss": 1.0792, + "step": 2795 + }, + { + "epoch": 0.33619912222689835, + "grad_norm": 2.292081437000608, + "learning_rate": 3.0946513016572464e-06, + "loss": 1.0294, + "step": 2796 + }, + { + "epoch": 0.3363193651175374, + "grad_norm": 2.310101579851981, + "learning_rate": 3.0939992819052938e-06, + "loss": 1.0131, + "step": 2797 + }, + { + "epoch": 0.3364396080081765, + "grad_norm": 2.2783030457379883, + "learning_rate": 3.0933470961945193e-06, + "loss": 1.0525, + "step": 2798 + }, + { + "epoch": 0.3365598508988156, + "grad_norm": 1.8076617680044247, + "learning_rate": 3.0926947446238597e-06, + "loss": 0.9184, + "step": 2799 + }, + { + "epoch": 0.3366800937894547, + "grad_norm": 2.1508135609221073, + "learning_rate": 3.092042227292276e-06, + "loss": 1.0615, + "step": 2800 + }, + { + "epoch": 0.3368003366800938, + "grad_norm": 1.733031568791092, + "learning_rate": 3.0913895442987557e-06, + "loss": 1.1207, + "step": 2801 + }, + { + "epoch": 0.3369205795707329, + "grad_norm": 1.7581767065186038, + "learning_rate": 3.090736695742308e-06, + "loss": 1.0957, + "step": 2802 + }, + { + "epoch": 0.33704082246137196, + "grad_norm": 2.3473923926832194, + "learning_rate": 3.0900836817219713e-06, + "loss": 0.762, + "step": 2803 + }, + { + "epoch": 0.33716106535201107, + "grad_norm": 1.7744827887032815, + "learning_rate": 3.089430502336807e-06, + "loss": 1.0717, + "step": 2804 + }, + { + "epoch": 0.3372813082426502, + "grad_norm": 3.3978318650363852, + "learning_rate": 3.088777157685902e-06, + "loss": 1.1415, + "step": 2805 + }, + { + "epoch": 0.33740155113328923, + "grad_norm": 2.4092722157127695, + "learning_rate": 3.088123647868367e-06, + "loss": 1.0907, + "step": 2806 + }, + { + "epoch": 0.33752179402392835, + "grad_norm": 2.01801660874548, + "learning_rate": 3.0874699729833405e-06, + "loss": 1.0498, + "step": 2807 + }, + { + "epoch": 0.3376420369145674, + "grad_norm": 1.7064702489200387, + "learning_rate": 3.086816133129983e-06, + "loss": 1.0366, + "step": 2808 + }, + { + "epoch": 0.3377622798052065, + "grad_norm": 1.9435505068541108, + "learning_rate": 3.0861621284074826e-06, + "loss": 0.9992, + "step": 2809 + }, + { + "epoch": 0.3378825226958456, + "grad_norm": 3.605128973824337, + "learning_rate": 3.085507958915051e-06, + "loss": 0.9675, + "step": 2810 + }, + { + "epoch": 0.3380027655864847, + "grad_norm": 2.0688422974802263, + "learning_rate": 3.084853624751925e-06, + "loss": 0.9498, + "step": 2811 + }, + { + "epoch": 0.3381230084771238, + "grad_norm": 2.017805945208689, + "learning_rate": 3.0841991260173668e-06, + "loss": 1.0971, + "step": 2812 + }, + { + "epoch": 0.3382432513677629, + "grad_norm": 2.0796460773060037, + "learning_rate": 3.0835444628106634e-06, + "loss": 1.0411, + "step": 2813 + }, + { + "epoch": 0.33836349425840195, + "grad_norm": 1.9544441956023597, + "learning_rate": 3.082889635231126e-06, + "loss": 1.0691, + "step": 2814 + }, + { + "epoch": 0.33848373714904106, + "grad_norm": 2.9140117732892166, + "learning_rate": 3.0822346433780925e-06, + "loss": 1.0075, + "step": 2815 + }, + { + "epoch": 0.3386039800396802, + "grad_norm": 1.955664242929496, + "learning_rate": 3.0815794873509237e-06, + "loss": 1.1109, + "step": 2816 + }, + { + "epoch": 0.33872422293031923, + "grad_norm": 1.8974285191678664, + "learning_rate": 3.0809241672490066e-06, + "loss": 0.9678, + "step": 2817 + }, + { + "epoch": 0.33884446582095834, + "grad_norm": 2.1326776818224213, + "learning_rate": 3.080268683171753e-06, + "loss": 1.0873, + "step": 2818 + }, + { + "epoch": 0.33896470871159745, + "grad_norm": 2.290891261843322, + "learning_rate": 3.0796130352185985e-06, + "loss": 1.1306, + "step": 2819 + }, + { + "epoch": 0.3390849516022365, + "grad_norm": 1.8439283123107477, + "learning_rate": 3.0789572234890057e-06, + "loss": 0.9057, + "step": 2820 + }, + { + "epoch": 0.3392051944928756, + "grad_norm": 1.588047672026999, + "learning_rate": 3.0783012480824596e-06, + "loss": 1.013, + "step": 2821 + }, + { + "epoch": 0.33932543738351467, + "grad_norm": 2.6841299529132296, + "learning_rate": 3.077645109098471e-06, + "loss": 0.9821, + "step": 2822 + }, + { + "epoch": 0.3394456802741538, + "grad_norm": 1.7941956820388285, + "learning_rate": 3.076988806636577e-06, + "loss": 0.9588, + "step": 2823 + }, + { + "epoch": 0.3395659231647929, + "grad_norm": 2.074511606120984, + "learning_rate": 3.0763323407963377e-06, + "loss": 1.128, + "step": 2824 + }, + { + "epoch": 0.33968616605543195, + "grad_norm": 2.0314971773436397, + "learning_rate": 3.075675711677337e-06, + "loss": 1.0398, + "step": 2825 + }, + { + "epoch": 0.33980640894607106, + "grad_norm": 1.8946725065131034, + "learning_rate": 3.0750189193791865e-06, + "loss": 1.0166, + "step": 2826 + }, + { + "epoch": 0.33992665183671017, + "grad_norm": 2.5655854420742834, + "learning_rate": 3.0743619640015203e-06, + "loss": 0.9425, + "step": 2827 + }, + { + "epoch": 0.3400468947273492, + "grad_norm": 2.0214883850352736, + "learning_rate": 3.073704845643999e-06, + "loss": 1.1638, + "step": 2828 + }, + { + "epoch": 0.34016713761798834, + "grad_norm": 4.095771348755223, + "learning_rate": 3.0730475644063063e-06, + "loss": 1.0153, + "step": 2829 + }, + { + "epoch": 0.34028738050862745, + "grad_norm": 1.7789221828653523, + "learning_rate": 3.072390120388151e-06, + "loss": 0.8876, + "step": 2830 + }, + { + "epoch": 0.3404076233992665, + "grad_norm": 1.9702005643457137, + "learning_rate": 3.071732513689267e-06, + "loss": 0.9517, + "step": 2831 + }, + { + "epoch": 0.3405278662899056, + "grad_norm": 2.3097713971658598, + "learning_rate": 3.0710747444094134e-06, + "loss": 0.9093, + "step": 2832 + }, + { + "epoch": 0.3406481091805447, + "grad_norm": 2.2397051637617515, + "learning_rate": 3.070416812648372e-06, + "loss": 0.8885, + "step": 2833 + }, + { + "epoch": 0.3407683520711838, + "grad_norm": 3.746313444662523, + "learning_rate": 3.069758718505951e-06, + "loss": 0.8939, + "step": 2834 + }, + { + "epoch": 0.3408885949618229, + "grad_norm": 1.8095581910684146, + "learning_rate": 3.0691004620819836e-06, + "loss": 1.0402, + "step": 2835 + }, + { + "epoch": 0.341008837852462, + "grad_norm": 0.8128813802183342, + "learning_rate": 3.0684420434763254e-06, + "loss": 0.8399, + "step": 2836 + }, + { + "epoch": 0.34112908074310105, + "grad_norm": 1.8638048376027756, + "learning_rate": 3.06778346278886e-06, + "loss": 1.0103, + "step": 2837 + }, + { + "epoch": 0.34124932363374016, + "grad_norm": 2.4266879587846275, + "learning_rate": 3.0671247201194906e-06, + "loss": 1.0285, + "step": 2838 + }, + { + "epoch": 0.3413695665243792, + "grad_norm": 1.8114778332997892, + "learning_rate": 3.066465815568151e-06, + "loss": 0.9925, + "step": 2839 + }, + { + "epoch": 0.34148980941501833, + "grad_norm": 2.4648660939892064, + "learning_rate": 3.0658067492347947e-06, + "loss": 0.924, + "step": 2840 + }, + { + "epoch": 0.34161005230565744, + "grad_norm": 2.1066030868561256, + "learning_rate": 3.065147521219402e-06, + "loss": 0.9076, + "step": 2841 + }, + { + "epoch": 0.3417302951962965, + "grad_norm": 2.5883623722442826, + "learning_rate": 3.064488131621977e-06, + "loss": 0.9816, + "step": 2842 + }, + { + "epoch": 0.3418505380869356, + "grad_norm": 2.0334784388228035, + "learning_rate": 3.063828580542549e-06, + "loss": 0.9757, + "step": 2843 + }, + { + "epoch": 0.3419707809775747, + "grad_norm": 2.092617354845082, + "learning_rate": 3.0631688680811706e-06, + "loss": 0.9659, + "step": 2844 + }, + { + "epoch": 0.3420910238682138, + "grad_norm": 2.1072104191513086, + "learning_rate": 3.062508994337921e-06, + "loss": 0.9962, + "step": 2845 + }, + { + "epoch": 0.3422112667588529, + "grad_norm": 2.40602348549559, + "learning_rate": 3.0618489594129013e-06, + "loss": 1.0324, + "step": 2846 + }, + { + "epoch": 0.342331509649492, + "grad_norm": 2.096162800712546, + "learning_rate": 3.061188763406239e-06, + "loss": 0.9484, + "step": 2847 + }, + { + "epoch": 0.34245175254013105, + "grad_norm": 2.5080875234040243, + "learning_rate": 3.060528406418085e-06, + "loss": 1.0591, + "step": 2848 + }, + { + "epoch": 0.34257199543077016, + "grad_norm": 1.6429878295429103, + "learning_rate": 3.0598678885486145e-06, + "loss": 0.8515, + "step": 2849 + }, + { + "epoch": 0.34269223832140927, + "grad_norm": 2.529802073098355, + "learning_rate": 3.0592072098980282e-06, + "loss": 0.983, + "step": 2850 + }, + { + "epoch": 0.3428124812120483, + "grad_norm": 1.983595687546747, + "learning_rate": 3.0585463705665514e-06, + "loss": 0.9662, + "step": 2851 + }, + { + "epoch": 0.34293272410268744, + "grad_norm": 2.2949744425495386, + "learning_rate": 3.0578853706544304e-06, + "loss": 0.9427, + "step": 2852 + }, + { + "epoch": 0.34305296699332655, + "grad_norm": 3.72660472534057, + "learning_rate": 3.0572242102619404e-06, + "loss": 0.8952, + "step": 2853 + }, + { + "epoch": 0.3431732098839656, + "grad_norm": 2.1392508223945415, + "learning_rate": 3.0565628894893784e-06, + "loss": 1.0486, + "step": 2854 + }, + { + "epoch": 0.3432934527746047, + "grad_norm": 1.8824073070588574, + "learning_rate": 3.0559014084370655e-06, + "loss": 0.9864, + "step": 2855 + }, + { + "epoch": 0.34341369566524377, + "grad_norm": 1.8515205861429829, + "learning_rate": 3.055239767205349e-06, + "loss": 1.0252, + "step": 2856 + }, + { + "epoch": 0.3435339385558829, + "grad_norm": 1.862808263473416, + "learning_rate": 3.054577965894599e-06, + "loss": 1.0219, + "step": 2857 + }, + { + "epoch": 0.343654181446522, + "grad_norm": 1.6396707575302962, + "learning_rate": 3.0539160046052094e-06, + "loss": 0.9402, + "step": 2858 + }, + { + "epoch": 0.34377442433716104, + "grad_norm": 2.950978301167252, + "learning_rate": 3.0532538834376003e-06, + "loss": 0.9434, + "step": 2859 + }, + { + "epoch": 0.34389466722780015, + "grad_norm": 2.024297700768207, + "learning_rate": 3.0525916024922143e-06, + "loss": 1.0204, + "step": 2860 + }, + { + "epoch": 0.34401491011843927, + "grad_norm": 3.358518269434527, + "learning_rate": 3.0519291618695193e-06, + "loss": 1.0754, + "step": 2861 + }, + { + "epoch": 0.3441351530090783, + "grad_norm": 1.7958657789138985, + "learning_rate": 3.0512665616700065e-06, + "loss": 0.9959, + "step": 2862 + }, + { + "epoch": 0.34425539589971743, + "grad_norm": 1.9701705530570024, + "learning_rate": 3.0506038019941933e-06, + "loss": 1.1304, + "step": 2863 + }, + { + "epoch": 0.34437563879035654, + "grad_norm": 2.646409239881315, + "learning_rate": 3.049940882942617e-06, + "loss": 0.9132, + "step": 2864 + }, + { + "epoch": 0.3444958816809956, + "grad_norm": 3.2181127713716213, + "learning_rate": 3.0492778046158448e-06, + "loss": 1.043, + "step": 2865 + }, + { + "epoch": 0.3446161245716347, + "grad_norm": 2.359481303870711, + "learning_rate": 3.0486145671144633e-06, + "loss": 1.0052, + "step": 2866 + }, + { + "epoch": 0.3447363674622738, + "grad_norm": 3.050042017662917, + "learning_rate": 3.047951170539086e-06, + "loss": 1.0033, + "step": 2867 + }, + { + "epoch": 0.3448566103529129, + "grad_norm": 1.9414795030862237, + "learning_rate": 3.047287614990349e-06, + "loss": 1.076, + "step": 2868 + }, + { + "epoch": 0.344976853243552, + "grad_norm": 2.3034108909691944, + "learning_rate": 3.046623900568914e-06, + "loss": 0.8592, + "step": 2869 + }, + { + "epoch": 0.34509709613419104, + "grad_norm": 2.607981509611193, + "learning_rate": 3.045960027375465e-06, + "loss": 0.9436, + "step": 2870 + }, + { + "epoch": 0.34521733902483015, + "grad_norm": 3.078477558529401, + "learning_rate": 3.045295995510711e-06, + "loss": 1.0635, + "step": 2871 + }, + { + "epoch": 0.34533758191546926, + "grad_norm": 1.7676561538587012, + "learning_rate": 3.0446318050753865e-06, + "loss": 0.9758, + "step": 2872 + }, + { + "epoch": 0.3454578248061083, + "grad_norm": 2.0285175431461178, + "learning_rate": 3.0439674561702474e-06, + "loss": 1.0095, + "step": 2873 + }, + { + "epoch": 0.3455780676967474, + "grad_norm": 2.7069113788789334, + "learning_rate": 3.043302948896076e-06, + "loss": 1.1203, + "step": 2874 + }, + { + "epoch": 0.34569831058738654, + "grad_norm": 2.262170828799871, + "learning_rate": 3.0426382833536756e-06, + "loss": 0.8412, + "step": 2875 + }, + { + "epoch": 0.3458185534780256, + "grad_norm": 2.021559384930791, + "learning_rate": 3.041973459643877e-06, + "loss": 1.0176, + "step": 2876 + }, + { + "epoch": 0.3459387963686647, + "grad_norm": 2.4282253563760627, + "learning_rate": 3.0413084778675334e-06, + "loss": 0.91, + "step": 2877 + }, + { + "epoch": 0.3460590392593038, + "grad_norm": 1.9694482685237908, + "learning_rate": 3.0406433381255214e-06, + "loss": 1.0793, + "step": 2878 + }, + { + "epoch": 0.34617928214994287, + "grad_norm": 4.402813809337986, + "learning_rate": 3.0399780405187425e-06, + "loss": 1.0579, + "step": 2879 + }, + { + "epoch": 0.346299525040582, + "grad_norm": 2.169772330967284, + "learning_rate": 3.0393125851481216e-06, + "loss": 1.0263, + "step": 2880 + }, + { + "epoch": 0.3464197679312211, + "grad_norm": 2.1769486040348713, + "learning_rate": 3.038646972114608e-06, + "loss": 1.1022, + "step": 2881 + }, + { + "epoch": 0.34654001082186014, + "grad_norm": 2.9130492711612996, + "learning_rate": 3.037981201519174e-06, + "loss": 0.9102, + "step": 2882 + }, + { + "epoch": 0.34666025371249926, + "grad_norm": 1.9611720214668635, + "learning_rate": 3.0373152734628175e-06, + "loss": 0.9523, + "step": 2883 + }, + { + "epoch": 0.34678049660313837, + "grad_norm": 2.0821242320642215, + "learning_rate": 3.0366491880465584e-06, + "loss": 1.0024, + "step": 2884 + }, + { + "epoch": 0.3469007394937774, + "grad_norm": 1.6427935845655923, + "learning_rate": 3.035982945371443e-06, + "loss": 1.0586, + "step": 2885 + }, + { + "epoch": 0.34702098238441653, + "grad_norm": 2.0579256135270163, + "learning_rate": 3.035316545538537e-06, + "loss": 1.0915, + "step": 2886 + }, + { + "epoch": 0.3471412252750556, + "grad_norm": 2.0695149738943304, + "learning_rate": 3.034649988648935e-06, + "loss": 1.0321, + "step": 2887 + }, + { + "epoch": 0.3472614681656947, + "grad_norm": 1.777606315920723, + "learning_rate": 3.033983274803752e-06, + "loss": 1.0431, + "step": 2888 + }, + { + "epoch": 0.3473817110563338, + "grad_norm": 2.203729749425323, + "learning_rate": 3.0333164041041283e-06, + "loss": 0.9596, + "step": 2889 + }, + { + "epoch": 0.34750195394697286, + "grad_norm": 2.0945694871177576, + "learning_rate": 3.032649376651228e-06, + "loss": 0.958, + "step": 2890 + }, + { + "epoch": 0.347622196837612, + "grad_norm": 1.6835193307807388, + "learning_rate": 3.031982192546238e-06, + "loss": 0.9933, + "step": 2891 + }, + { + "epoch": 0.3477424397282511, + "grad_norm": 2.896409633749945, + "learning_rate": 3.0313148518903696e-06, + "loss": 1.1812, + "step": 2892 + }, + { + "epoch": 0.34786268261889014, + "grad_norm": 2.4307324287605723, + "learning_rate": 3.030647354784859e-06, + "loss": 1.0484, + "step": 2893 + }, + { + "epoch": 0.34798292550952925, + "grad_norm": 1.9012293753423848, + "learning_rate": 3.029979701330964e-06, + "loss": 1.0102, + "step": 2894 + }, + { + "epoch": 0.34810316840016836, + "grad_norm": 2.546637120283533, + "learning_rate": 3.029311891629966e-06, + "loss": 1.0373, + "step": 2895 + }, + { + "epoch": 0.3482234112908074, + "grad_norm": 2.3459686967496287, + "learning_rate": 3.0286439257831744e-06, + "loss": 0.9821, + "step": 2896 + }, + { + "epoch": 0.3483436541814465, + "grad_norm": 1.9580280255346951, + "learning_rate": 3.0279758038919156e-06, + "loss": 0.9509, + "step": 2897 + }, + { + "epoch": 0.34846389707208564, + "grad_norm": 3.330944460943786, + "learning_rate": 3.0273075260575455e-06, + "loss": 1.0226, + "step": 2898 + }, + { + "epoch": 0.3485841399627247, + "grad_norm": 2.3940359477672133, + "learning_rate": 3.0266390923814396e-06, + "loss": 1.0468, + "step": 2899 + }, + { + "epoch": 0.3487043828533638, + "grad_norm": 1.8609568815078825, + "learning_rate": 3.0259705029650008e-06, + "loss": 1.0618, + "step": 2900 + }, + { + "epoch": 0.34882462574400286, + "grad_norm": 1.7877354260110518, + "learning_rate": 3.025301757909652e-06, + "loss": 0.9669, + "step": 2901 + }, + { + "epoch": 0.34894486863464197, + "grad_norm": 1.6015102470991938, + "learning_rate": 3.024632857316842e-06, + "loss": 1.0447, + "step": 2902 + }, + { + "epoch": 0.3490651115252811, + "grad_norm": 2.5702235429669407, + "learning_rate": 3.0239638012880412e-06, + "loss": 1.016, + "step": 2903 + }, + { + "epoch": 0.34918535441592014, + "grad_norm": 2.552598945276607, + "learning_rate": 3.0232945899247466e-06, + "loss": 1.0456, + "step": 2904 + }, + { + "epoch": 0.34930559730655925, + "grad_norm": 2.2227931190009405, + "learning_rate": 3.022625223328476e-06, + "loss": 1.0094, + "step": 2905 + }, + { + "epoch": 0.34942584019719836, + "grad_norm": 1.5994808904633777, + "learning_rate": 3.0219557016007723e-06, + "loss": 0.928, + "step": 2906 + }, + { + "epoch": 0.3495460830878374, + "grad_norm": 1.9865331853384192, + "learning_rate": 3.021286024843202e-06, + "loss": 0.9358, + "step": 2907 + }, + { + "epoch": 0.3496663259784765, + "grad_norm": 1.0809649229472407, + "learning_rate": 3.0206161931573526e-06, + "loss": 0.8862, + "step": 2908 + }, + { + "epoch": 0.34978656886911563, + "grad_norm": 1.9790860075994863, + "learning_rate": 3.0199462066448388e-06, + "loss": 1.1702, + "step": 2909 + }, + { + "epoch": 0.3499068117597547, + "grad_norm": 1.929915132880262, + "learning_rate": 3.019276065407296e-06, + "loss": 0.9313, + "step": 2910 + }, + { + "epoch": 0.3500270546503938, + "grad_norm": 1.7497995806778224, + "learning_rate": 3.018605769546385e-06, + "loss": 1.047, + "step": 2911 + }, + { + "epoch": 0.3501472975410329, + "grad_norm": 1.9771540664974079, + "learning_rate": 3.017935319163788e-06, + "loss": 1.0339, + "step": 2912 + }, + { + "epoch": 0.35026754043167196, + "grad_norm": 1.9325848654289315, + "learning_rate": 3.017264714361213e-06, + "loss": 0.9426, + "step": 2913 + }, + { + "epoch": 0.3503877833223111, + "grad_norm": 2.124304489307843, + "learning_rate": 3.016593955240389e-06, + "loss": 1.0576, + "step": 2914 + }, + { + "epoch": 0.3505080262129502, + "grad_norm": 0.8307924675018243, + "learning_rate": 3.015923041903071e-06, + "loss": 0.8749, + "step": 2915 + }, + { + "epoch": 0.35062826910358924, + "grad_norm": 2.1393589257436862, + "learning_rate": 3.0152519744510347e-06, + "loss": 1.0695, + "step": 2916 + }, + { + "epoch": 0.35074851199422835, + "grad_norm": 1.8796033299754524, + "learning_rate": 3.014580752986081e-06, + "loss": 1.0673, + "step": 2917 + }, + { + "epoch": 0.3508687548848674, + "grad_norm": 1.9208654217580834, + "learning_rate": 3.0139093776100345e-06, + "loss": 1.0228, + "step": 2918 + }, + { + "epoch": 0.3509889977755065, + "grad_norm": 1.765137866852151, + "learning_rate": 3.013237848424741e-06, + "loss": 0.9882, + "step": 2919 + }, + { + "epoch": 0.35110924066614563, + "grad_norm": 2.1687224856343525, + "learning_rate": 3.012566165532072e-06, + "loss": 0.9937, + "step": 2920 + }, + { + "epoch": 0.3512294835567847, + "grad_norm": 1.964465671817067, + "learning_rate": 3.0118943290339207e-06, + "loss": 1.0039, + "step": 2921 + }, + { + "epoch": 0.3513497264474238, + "grad_norm": 1.9304068147487634, + "learning_rate": 3.011222339032204e-06, + "loss": 0.9132, + "step": 2922 + }, + { + "epoch": 0.3514699693380629, + "grad_norm": 2.1878900452694623, + "learning_rate": 3.0105501956288626e-06, + "loss": 0.934, + "step": 2923 + }, + { + "epoch": 0.35159021222870196, + "grad_norm": 2.0627200847301648, + "learning_rate": 3.0098778989258602e-06, + "loss": 0.9656, + "step": 2924 + }, + { + "epoch": 0.35171045511934107, + "grad_norm": 2.541055595441299, + "learning_rate": 3.009205449025183e-06, + "loss": 1.1184, + "step": 2925 + }, + { + "epoch": 0.3518306980099802, + "grad_norm": 1.882879865398856, + "learning_rate": 3.008532846028842e-06, + "loss": 0.8665, + "step": 2926 + }, + { + "epoch": 0.35195094090061924, + "grad_norm": 12.010778672562786, + "learning_rate": 3.0078600900388694e-06, + "loss": 0.9545, + "step": 2927 + }, + { + "epoch": 0.35207118379125835, + "grad_norm": 1.7869688885611028, + "learning_rate": 3.007187181157323e-06, + "loss": 0.9778, + "step": 2928 + }, + { + "epoch": 0.35219142668189746, + "grad_norm": 2.454645742429077, + "learning_rate": 3.006514119486282e-06, + "loss": 0.9202, + "step": 2929 + }, + { + "epoch": 0.3523116695725365, + "grad_norm": 2.2234884892914826, + "learning_rate": 3.005840905127849e-06, + "loss": 0.9362, + "step": 2930 + }, + { + "epoch": 0.3524319124631756, + "grad_norm": 2.2399783000576154, + "learning_rate": 3.0051675381841516e-06, + "loss": 1.1047, + "step": 2931 + }, + { + "epoch": 0.3525521553538147, + "grad_norm": 1.656138866222838, + "learning_rate": 3.0044940187573363e-06, + "loss": 1.0068, + "step": 2932 + }, + { + "epoch": 0.3526723982444538, + "grad_norm": 2.19907987166267, + "learning_rate": 3.003820346949578e-06, + "loss": 0.8932, + "step": 2933 + }, + { + "epoch": 0.3527926411350929, + "grad_norm": 2.030127934337896, + "learning_rate": 3.003146522863071e-06, + "loss": 1.0328, + "step": 2934 + }, + { + "epoch": 0.35291288402573195, + "grad_norm": 2.1216873827589895, + "learning_rate": 3.0024725466000345e-06, + "loss": 1.0949, + "step": 2935 + }, + { + "epoch": 0.35303312691637107, + "grad_norm": 1.9183459872378885, + "learning_rate": 3.0017984182627087e-06, + "loss": 1.0249, + "step": 2936 + }, + { + "epoch": 0.3531533698070102, + "grad_norm": 2.0091865083317804, + "learning_rate": 3.00112413795336e-06, + "loss": 1.0603, + "step": 2937 + }, + { + "epoch": 0.35327361269764923, + "grad_norm": 1.9730381359933173, + "learning_rate": 3.000449705774275e-06, + "loss": 1.0413, + "step": 2938 + }, + { + "epoch": 0.35339385558828834, + "grad_norm": 2.4222247455692663, + "learning_rate": 2.9997751218277654e-06, + "loss": 0.9532, + "step": 2939 + }, + { + "epoch": 0.35351409847892745, + "grad_norm": 7.872960628868959, + "learning_rate": 2.999100386216166e-06, + "loss": 1.0175, + "step": 2940 + }, + { + "epoch": 0.3536343413695665, + "grad_norm": 1.8906644706704354, + "learning_rate": 2.998425499041831e-06, + "loss": 0.9803, + "step": 2941 + }, + { + "epoch": 0.3537545842602056, + "grad_norm": 1.2685920880143882, + "learning_rate": 2.997750460407142e-06, + "loss": 0.8834, + "step": 2942 + }, + { + "epoch": 0.35387482715084473, + "grad_norm": 2.295335526217786, + "learning_rate": 2.997075270414501e-06, + "loss": 0.9369, + "step": 2943 + }, + { + "epoch": 0.3539950700414838, + "grad_norm": 0.707710150925132, + "learning_rate": 2.9963999291663347e-06, + "loss": 0.8153, + "step": 2944 + }, + { + "epoch": 0.3541153129321229, + "grad_norm": 2.6284355206924874, + "learning_rate": 2.9957244367650915e-06, + "loss": 0.9733, + "step": 2945 + }, + { + "epoch": 0.354235555822762, + "grad_norm": 2.0740068977631183, + "learning_rate": 2.9950487933132425e-06, + "loss": 1.0806, + "step": 2946 + }, + { + "epoch": 0.35435579871340106, + "grad_norm": 3.4712217305192707, + "learning_rate": 2.994372998913283e-06, + "loss": 0.949, + "step": 2947 + }, + { + "epoch": 0.35447604160404017, + "grad_norm": 4.936205537334474, + "learning_rate": 2.99369705366773e-06, + "loss": 0.8622, + "step": 2948 + }, + { + "epoch": 0.3545962844946792, + "grad_norm": 1.9769281946165678, + "learning_rate": 2.9930209576791244e-06, + "loss": 1.0593, + "step": 2949 + }, + { + "epoch": 0.35471652738531834, + "grad_norm": 2.882465620803611, + "learning_rate": 2.9923447110500285e-06, + "loss": 0.8726, + "step": 2950 + }, + { + "epoch": 0.35483677027595745, + "grad_norm": 1.488515439969252, + "learning_rate": 2.9916683138830295e-06, + "loss": 0.9931, + "step": 2951 + }, + { + "epoch": 0.3549570131665965, + "grad_norm": 2.4650537368147822, + "learning_rate": 2.9909917662807353e-06, + "loss": 1.0459, + "step": 2952 + }, + { + "epoch": 0.3550772560572356, + "grad_norm": 2.3292001379493543, + "learning_rate": 2.9903150683457783e-06, + "loss": 0.9326, + "step": 2953 + }, + { + "epoch": 0.3551974989478747, + "grad_norm": 2.5169943168857873, + "learning_rate": 2.9896382201808126e-06, + "loss": 0.8887, + "step": 2954 + }, + { + "epoch": 0.3553177418385138, + "grad_norm": 2.9528735825826264, + "learning_rate": 2.988961221888516e-06, + "loss": 1.0441, + "step": 2955 + }, + { + "epoch": 0.3554379847291529, + "grad_norm": 2.5216429384993124, + "learning_rate": 2.988284073571589e-06, + "loss": 1.0275, + "step": 2956 + }, + { + "epoch": 0.355558227619792, + "grad_norm": 14.785380199757029, + "learning_rate": 2.9876067753327528e-06, + "loss": 0.9675, + "step": 2957 + }, + { + "epoch": 0.35567847051043106, + "grad_norm": 2.0684092860553234, + "learning_rate": 2.986929327274754e-06, + "loss": 1.0475, + "step": 2958 + }, + { + "epoch": 0.35579871340107017, + "grad_norm": 1.6877550003097477, + "learning_rate": 2.9862517295003617e-06, + "loss": 1.0217, + "step": 2959 + }, + { + "epoch": 0.3559189562917093, + "grad_norm": 1.711141217092132, + "learning_rate": 2.9855739821123654e-06, + "loss": 0.964, + "step": 2960 + }, + { + "epoch": 0.35603919918234833, + "grad_norm": 1.7817961541496903, + "learning_rate": 2.98489608521358e-06, + "loss": 1.0597, + "step": 2961 + }, + { + "epoch": 0.35615944207298744, + "grad_norm": 2.8583415514176305, + "learning_rate": 2.9842180389068425e-06, + "loss": 1.0383, + "step": 2962 + }, + { + "epoch": 0.35627968496362655, + "grad_norm": 0.7707492464614337, + "learning_rate": 2.98353984329501e-06, + "loss": 0.8294, + "step": 2963 + }, + { + "epoch": 0.3563999278542656, + "grad_norm": 1.7420865016793787, + "learning_rate": 2.982861498480965e-06, + "loss": 0.9441, + "step": 2964 + }, + { + "epoch": 0.3565201707449047, + "grad_norm": 3.4160474510738696, + "learning_rate": 2.9821830045676122e-06, + "loss": 1.0637, + "step": 2965 + }, + { + "epoch": 0.3566404136355438, + "grad_norm": 1.6875091835054656, + "learning_rate": 2.9815043616578793e-06, + "loss": 0.9694, + "step": 2966 + }, + { + "epoch": 0.3567606565261829, + "grad_norm": 2.223872651378572, + "learning_rate": 2.9808255698547145e-06, + "loss": 1.0075, + "step": 2967 + }, + { + "epoch": 0.356880899416822, + "grad_norm": 2.2069486079314435, + "learning_rate": 2.9801466292610913e-06, + "loss": 1.0391, + "step": 2968 + }, + { + "epoch": 0.35700114230746105, + "grad_norm": 2.3174770511285163, + "learning_rate": 2.979467539980003e-06, + "loss": 1.0494, + "step": 2969 + }, + { + "epoch": 0.35712138519810016, + "grad_norm": 2.0394708801794286, + "learning_rate": 2.978788302114468e-06, + "loss": 1.0094, + "step": 2970 + }, + { + "epoch": 0.35724162808873927, + "grad_norm": 2.12827811117482, + "learning_rate": 2.9781089157675255e-06, + "loss": 1.0705, + "step": 2971 + }, + { + "epoch": 0.3573618709793783, + "grad_norm": 1.6928448274795644, + "learning_rate": 2.977429381042238e-06, + "loss": 1.1249, + "step": 2972 + }, + { + "epoch": 0.35748211387001744, + "grad_norm": 2.744511328803261, + "learning_rate": 2.9767496980416913e-06, + "loss": 1.125, + "step": 2973 + }, + { + "epoch": 0.35760235676065655, + "grad_norm": 4.145433850746275, + "learning_rate": 2.9760698668689914e-06, + "loss": 1.0492, + "step": 2974 + }, + { + "epoch": 0.3577225996512956, + "grad_norm": 1.9587338896906232, + "learning_rate": 2.975389887627269e-06, + "loss": 0.9506, + "step": 2975 + }, + { + "epoch": 0.3578428425419347, + "grad_norm": 2.316080980799061, + "learning_rate": 2.9747097604196764e-06, + "loss": 1.1375, + "step": 2976 + }, + { + "epoch": 0.3579630854325738, + "grad_norm": 0.6754040188892614, + "learning_rate": 2.9740294853493875e-06, + "loss": 0.8039, + "step": 2977 + }, + { + "epoch": 0.3580833283232129, + "grad_norm": 1.8806993331240267, + "learning_rate": 2.9733490625196008e-06, + "loss": 0.9077, + "step": 2978 + }, + { + "epoch": 0.358203571213852, + "grad_norm": 3.586468027799594, + "learning_rate": 2.9726684920335353e-06, + "loss": 0.9938, + "step": 2979 + }, + { + "epoch": 0.35832381410449105, + "grad_norm": 4.364882186575664, + "learning_rate": 2.971987773994432e-06, + "loss": 1.0585, + "step": 2980 + }, + { + "epoch": 0.35844405699513016, + "grad_norm": 2.7253352247755402, + "learning_rate": 2.9713069085055566e-06, + "loss": 1.0675, + "step": 2981 + }, + { + "epoch": 0.35856429988576927, + "grad_norm": 2.223311342800726, + "learning_rate": 2.9706258956701958e-06, + "loss": 1.0277, + "step": 2982 + }, + { + "epoch": 0.3586845427764083, + "grad_norm": 2.8136812060846332, + "learning_rate": 2.9699447355916575e-06, + "loss": 1.0108, + "step": 2983 + }, + { + "epoch": 0.35880478566704743, + "grad_norm": 1.9544605419150713, + "learning_rate": 2.969263428373275e-06, + "loss": 0.97, + "step": 2984 + }, + { + "epoch": 0.35892502855768654, + "grad_norm": 5.120036815570553, + "learning_rate": 2.9685819741184007e-06, + "loss": 1.0312, + "step": 2985 + }, + { + "epoch": 0.3590452714483256, + "grad_norm": 2.442959369266897, + "learning_rate": 2.967900372930411e-06, + "loss": 0.922, + "step": 2986 + }, + { + "epoch": 0.3591655143389647, + "grad_norm": 2.9345204277686006, + "learning_rate": 2.9672186249127046e-06, + "loss": 1.0298, + "step": 2987 + }, + { + "epoch": 0.3592857572296038, + "grad_norm": 1.995080082081992, + "learning_rate": 2.9665367301687014e-06, + "loss": 1.0297, + "step": 2988 + }, + { + "epoch": 0.3594060001202429, + "grad_norm": 2.423736054103445, + "learning_rate": 2.965854688801845e-06, + "loss": 1.0025, + "step": 2989 + }, + { + "epoch": 0.359526243010882, + "grad_norm": 2.1773378019658844, + "learning_rate": 2.9651725009156005e-06, + "loss": 1.0022, + "step": 2990 + }, + { + "epoch": 0.3596464859015211, + "grad_norm": 1.8645451575549914, + "learning_rate": 2.964490166613454e-06, + "loss": 0.9799, + "step": 2991 + }, + { + "epoch": 0.35976672879216015, + "grad_norm": 0.7711355072433532, + "learning_rate": 2.963807685998917e-06, + "loss": 0.8146, + "step": 2992 + }, + { + "epoch": 0.35988697168279926, + "grad_norm": 1.5588215826342045, + "learning_rate": 2.9631250591755196e-06, + "loss": 1.0165, + "step": 2993 + }, + { + "epoch": 0.36000721457343837, + "grad_norm": 1.6988342551697284, + "learning_rate": 2.962442286246817e-06, + "loss": 0.8154, + "step": 2994 + }, + { + "epoch": 0.3601274574640774, + "grad_norm": 1.8301096519596307, + "learning_rate": 2.9617593673163853e-06, + "loss": 0.937, + "step": 2995 + }, + { + "epoch": 0.36024770035471654, + "grad_norm": 2.0719837281018987, + "learning_rate": 2.9610763024878216e-06, + "loss": 1.0125, + "step": 2996 + }, + { + "epoch": 0.3603679432453556, + "grad_norm": 1.7797644594347013, + "learning_rate": 2.960393091864747e-06, + "loss": 1.1533, + "step": 2997 + }, + { + "epoch": 0.3604881861359947, + "grad_norm": 2.0061242988162586, + "learning_rate": 2.959709735550804e-06, + "loss": 0.9869, + "step": 2998 + }, + { + "epoch": 0.3606084290266338, + "grad_norm": 2.112944796532981, + "learning_rate": 2.9590262336496575e-06, + "loss": 0.9956, + "step": 2999 + }, + { + "epoch": 0.36072867191727287, + "grad_norm": 2.9510927019285544, + "learning_rate": 2.9583425862649936e-06, + "loss": 1.0941, + "step": 3000 + }, + { + "epoch": 0.360848914807912, + "grad_norm": 2.4816084518760917, + "learning_rate": 2.9576587935005215e-06, + "loss": 0.9773, + "step": 3001 + }, + { + "epoch": 0.3609691576985511, + "grad_norm": 2.857080925542729, + "learning_rate": 2.9569748554599713e-06, + "loss": 0.9586, + "step": 3002 + }, + { + "epoch": 0.36108940058919015, + "grad_norm": 2.163741656044354, + "learning_rate": 2.956290772247097e-06, + "loss": 0.9701, + "step": 3003 + }, + { + "epoch": 0.36120964347982926, + "grad_norm": 1.9015444852654275, + "learning_rate": 2.9556065439656724e-06, + "loss": 0.9721, + "step": 3004 + }, + { + "epoch": 0.36132988637046837, + "grad_norm": 1.6980400820697819, + "learning_rate": 2.9549221707194952e-06, + "loss": 1.0523, + "step": 3005 + }, + { + "epoch": 0.3614501292611074, + "grad_norm": 2.0756001443359655, + "learning_rate": 2.954237652612384e-06, + "loss": 0.967, + "step": 3006 + }, + { + "epoch": 0.36157037215174653, + "grad_norm": 2.2988628686241843, + "learning_rate": 2.9535529897481796e-06, + "loss": 1.0833, + "step": 3007 + }, + { + "epoch": 0.36169061504238564, + "grad_norm": 2.2084122649667712, + "learning_rate": 2.9528681822307446e-06, + "loss": 1.0084, + "step": 3008 + }, + { + "epoch": 0.3618108579330247, + "grad_norm": 2.2319416587615444, + "learning_rate": 2.952183230163964e-06, + "loss": 1.0596, + "step": 3009 + }, + { + "epoch": 0.3619311008236638, + "grad_norm": 2.022695869420234, + "learning_rate": 2.9514981336517448e-06, + "loss": 0.9685, + "step": 3010 + }, + { + "epoch": 0.36205134371430286, + "grad_norm": 1.8849434724219383, + "learning_rate": 2.950812892798015e-06, + "loss": 1.051, + "step": 3011 + }, + { + "epoch": 0.362171586604942, + "grad_norm": 1.992150579372188, + "learning_rate": 2.9501275077067256e-06, + "loss": 1.1105, + "step": 3012 + }, + { + "epoch": 0.3622918294955811, + "grad_norm": 1.6418478631781794, + "learning_rate": 2.949441978481848e-06, + "loss": 1.1176, + "step": 3013 + }, + { + "epoch": 0.36241207238622014, + "grad_norm": 2.36740614885294, + "learning_rate": 2.9487563052273778e-06, + "loss": 1.0387, + "step": 3014 + }, + { + "epoch": 0.36253231527685925, + "grad_norm": 2.260571968146324, + "learning_rate": 2.94807048804733e-06, + "loss": 1.0917, + "step": 3015 + }, + { + "epoch": 0.36265255816749836, + "grad_norm": 2.1009786142153226, + "learning_rate": 2.9473845270457434e-06, + "loss": 1.1445, + "step": 3016 + }, + { + "epoch": 0.3627728010581374, + "grad_norm": 2.1994990818917555, + "learning_rate": 2.946698422326677e-06, + "loss": 0.9343, + "step": 3017 + }, + { + "epoch": 0.36289304394877653, + "grad_norm": 2.1555477602097826, + "learning_rate": 2.946012173994213e-06, + "loss": 1.0335, + "step": 3018 + }, + { + "epoch": 0.36301328683941564, + "grad_norm": 2.2173809072001966, + "learning_rate": 2.945325782152454e-06, + "loss": 0.9084, + "step": 3019 + }, + { + "epoch": 0.3631335297300547, + "grad_norm": 2.0972240720267767, + "learning_rate": 2.9446392469055257e-06, + "loss": 1.021, + "step": 3020 + }, + { + "epoch": 0.3632537726206938, + "grad_norm": 2.2285783175477247, + "learning_rate": 2.9439525683575745e-06, + "loss": 1.0347, + "step": 3021 + }, + { + "epoch": 0.3633740155113329, + "grad_norm": 2.7929855666287873, + "learning_rate": 2.9432657466127694e-06, + "loss": 0.9931, + "step": 3022 + }, + { + "epoch": 0.36349425840197197, + "grad_norm": 1.6407233799093324, + "learning_rate": 2.9425787817753007e-06, + "loss": 1.0021, + "step": 3023 + }, + { + "epoch": 0.3636145012926111, + "grad_norm": 1.9181824854455565, + "learning_rate": 2.94189167394938e-06, + "loss": 0.9501, + "step": 3024 + }, + { + "epoch": 0.3637347441832502, + "grad_norm": 2.1517472730096108, + "learning_rate": 2.941204423239241e-06, + "loss": 1.0485, + "step": 3025 + }, + { + "epoch": 0.36385498707388925, + "grad_norm": 1.8413471146071747, + "learning_rate": 2.9405170297491395e-06, + "loss": 0.9954, + "step": 3026 + }, + { + "epoch": 0.36397522996452836, + "grad_norm": 1.9905201834925246, + "learning_rate": 2.939829493583353e-06, + "loss": 1.041, + "step": 3027 + }, + { + "epoch": 0.3640954728551674, + "grad_norm": 2.849085738118784, + "learning_rate": 2.939141814846179e-06, + "loss": 1.0723, + "step": 3028 + }, + { + "epoch": 0.3642157157458065, + "grad_norm": 3.9777481320115267, + "learning_rate": 2.938453993641938e-06, + "loss": 1.059, + "step": 3029 + }, + { + "epoch": 0.36433595863644563, + "grad_norm": 2.393578792654528, + "learning_rate": 2.937766030074973e-06, + "loss": 0.9403, + "step": 3030 + }, + { + "epoch": 0.3644562015270847, + "grad_norm": 1.9376556580066353, + "learning_rate": 2.937077924249646e-06, + "loss": 1.0657, + "step": 3031 + }, + { + "epoch": 0.3645764444177238, + "grad_norm": 1.9970655371275425, + "learning_rate": 2.9363896762703443e-06, + "loss": 0.9927, + "step": 3032 + }, + { + "epoch": 0.3646966873083629, + "grad_norm": 1.8176507439731209, + "learning_rate": 2.9357012862414725e-06, + "loss": 1.0814, + "step": 3033 + }, + { + "epoch": 0.36481693019900197, + "grad_norm": 1.83720140171907, + "learning_rate": 2.9350127542674593e-06, + "loss": 0.9543, + "step": 3034 + }, + { + "epoch": 0.3649371730896411, + "grad_norm": 1.9586727849156276, + "learning_rate": 2.934324080452755e-06, + "loss": 1.0084, + "step": 3035 + }, + { + "epoch": 0.3650574159802802, + "grad_norm": 1.5063897570595566, + "learning_rate": 2.9336352649018307e-06, + "loss": 1.0194, + "step": 3036 + }, + { + "epoch": 0.36517765887091924, + "grad_norm": 1.7499232956561073, + "learning_rate": 2.9329463077191783e-06, + "loss": 0.9388, + "step": 3037 + }, + { + "epoch": 0.36529790176155835, + "grad_norm": 1.953089965506094, + "learning_rate": 2.9322572090093135e-06, + "loss": 0.8773, + "step": 3038 + }, + { + "epoch": 0.36541814465219746, + "grad_norm": 3.061280922661488, + "learning_rate": 2.9315679688767713e-06, + "loss": 1.0087, + "step": 3039 + }, + { + "epoch": 0.3655383875428365, + "grad_norm": 1.8819450184475945, + "learning_rate": 2.9308785874261085e-06, + "loss": 0.9066, + "step": 3040 + }, + { + "epoch": 0.36565863043347563, + "grad_norm": 1.6858143334568096, + "learning_rate": 2.9301890647619045e-06, + "loss": 1.056, + "step": 3041 + }, + { + "epoch": 0.36577887332411474, + "grad_norm": 2.2907467007020537, + "learning_rate": 2.929499400988759e-06, + "loss": 1.0423, + "step": 3042 + }, + { + "epoch": 0.3658991162147538, + "grad_norm": 2.109183295686908, + "learning_rate": 2.9288095962112927e-06, + "loss": 0.8916, + "step": 3043 + }, + { + "epoch": 0.3660193591053929, + "grad_norm": 1.7987717941692267, + "learning_rate": 2.9281196505341503e-06, + "loss": 1.0856, + "step": 3044 + }, + { + "epoch": 0.36613960199603196, + "grad_norm": 2.0343193925561875, + "learning_rate": 2.9274295640619946e-06, + "loss": 1.0257, + "step": 3045 + }, + { + "epoch": 0.36625984488667107, + "grad_norm": 2.031204457872869, + "learning_rate": 2.9267393368995103e-06, + "loss": 1.0205, + "step": 3046 + }, + { + "epoch": 0.3663800877773102, + "grad_norm": 2.685277241435752, + "learning_rate": 2.926048969151407e-06, + "loss": 0.9824, + "step": 3047 + }, + { + "epoch": 0.36650033066794924, + "grad_norm": 1.8829302612453227, + "learning_rate": 2.92535846092241e-06, + "loss": 0.9207, + "step": 3048 + }, + { + "epoch": 0.36662057355858835, + "grad_norm": 2.0909637587264913, + "learning_rate": 2.9246678123172704e-06, + "loss": 1.0647, + "step": 3049 + }, + { + "epoch": 0.36674081644922746, + "grad_norm": 3.3732063330866926, + "learning_rate": 2.9239770234407596e-06, + "loss": 0.986, + "step": 3050 + }, + { + "epoch": 0.3668610593398665, + "grad_norm": 2.351105807241712, + "learning_rate": 2.9232860943976686e-06, + "loss": 0.9236, + "step": 3051 + }, + { + "epoch": 0.3669813022305056, + "grad_norm": 1.7097200564961164, + "learning_rate": 2.9225950252928115e-06, + "loss": 1.0787, + "step": 3052 + }, + { + "epoch": 0.36710154512114473, + "grad_norm": 2.089434717382169, + "learning_rate": 2.9219038162310217e-06, + "loss": 1.0594, + "step": 3053 + }, + { + "epoch": 0.3672217880117838, + "grad_norm": 2.189816365664489, + "learning_rate": 2.921212467317157e-06, + "loss": 1.0647, + "step": 3054 + }, + { + "epoch": 0.3673420309024229, + "grad_norm": 2.027393377207498, + "learning_rate": 2.920520978656093e-06, + "loss": 1.0394, + "step": 3055 + }, + { + "epoch": 0.367462273793062, + "grad_norm": 2.0307999332370112, + "learning_rate": 2.919829350352729e-06, + "loss": 1.0036, + "step": 3056 + }, + { + "epoch": 0.36758251668370107, + "grad_norm": 0.7781918753569206, + "learning_rate": 2.919137582511983e-06, + "loss": 0.8372, + "step": 3057 + }, + { + "epoch": 0.3677027595743402, + "grad_norm": 2.047518330586301, + "learning_rate": 2.918445675238797e-06, + "loss": 0.8743, + "step": 3058 + }, + { + "epoch": 0.36782300246497923, + "grad_norm": 1.9955039748742707, + "learning_rate": 2.917753628638132e-06, + "loss": 0.9381, + "step": 3059 + }, + { + "epoch": 0.36794324535561834, + "grad_norm": 2.3073741705777406, + "learning_rate": 2.9170614428149716e-06, + "loss": 0.944, + "step": 3060 + }, + { + "epoch": 0.36806348824625745, + "grad_norm": 2.810287877198397, + "learning_rate": 2.9163691178743195e-06, + "loss": 1.1079, + "step": 3061 + }, + { + "epoch": 0.3681837311368965, + "grad_norm": 2.0550729368311482, + "learning_rate": 2.9156766539212006e-06, + "loss": 1.0155, + "step": 3062 + }, + { + "epoch": 0.3683039740275356, + "grad_norm": 2.043023132417322, + "learning_rate": 2.9149840510606614e-06, + "loss": 0.9562, + "step": 3063 + }, + { + "epoch": 0.36842421691817473, + "grad_norm": 1.0322740845220646, + "learning_rate": 2.914291309397769e-06, + "loss": 0.8801, + "step": 3064 + }, + { + "epoch": 0.3685444598088138, + "grad_norm": 2.026431663130873, + "learning_rate": 2.9135984290376117e-06, + "loss": 1.0215, + "step": 3065 + }, + { + "epoch": 0.3686647026994529, + "grad_norm": 1.643652892054562, + "learning_rate": 2.9129054100853e-06, + "loss": 1.061, + "step": 3066 + }, + { + "epoch": 0.368784945590092, + "grad_norm": 1.782889498386033, + "learning_rate": 2.912212252645963e-06, + "loss": 0.994, + "step": 3067 + }, + { + "epoch": 0.36890518848073106, + "grad_norm": 2.335008723632723, + "learning_rate": 2.9115189568247523e-06, + "loss": 0.9995, + "step": 3068 + }, + { + "epoch": 0.36902543137137017, + "grad_norm": 2.026977035490236, + "learning_rate": 2.910825522726841e-06, + "loss": 1.1604, + "step": 3069 + }, + { + "epoch": 0.3691456742620093, + "grad_norm": 2.1789860297390278, + "learning_rate": 2.9101319504574215e-06, + "loss": 1.0071, + "step": 3070 + }, + { + "epoch": 0.36926591715264834, + "grad_norm": 1.9929819953744485, + "learning_rate": 2.909438240121709e-06, + "loss": 1.002, + "step": 3071 + }, + { + "epoch": 0.36938616004328745, + "grad_norm": 1.736849694714551, + "learning_rate": 2.908744391824939e-06, + "loss": 0.9449, + "step": 3072 + }, + { + "epoch": 0.36950640293392656, + "grad_norm": 1.9865729107058143, + "learning_rate": 2.908050405672367e-06, + "loss": 1.0307, + "step": 3073 + }, + { + "epoch": 0.3696266458245656, + "grad_norm": 1.8222834840739615, + "learning_rate": 2.9073562817692703e-06, + "loss": 1.028, + "step": 3074 + }, + { + "epoch": 0.3697468887152047, + "grad_norm": 0.7303520966267911, + "learning_rate": 2.9066620202209468e-06, + "loss": 0.8027, + "step": 3075 + }, + { + "epoch": 0.3698671316058438, + "grad_norm": 1.983797725302908, + "learning_rate": 2.905967621132716e-06, + "loss": 1.0171, + "step": 3076 + }, + { + "epoch": 0.3699873744964829, + "grad_norm": 2.0707507551321505, + "learning_rate": 2.9052730846099172e-06, + "loss": 0.9868, + "step": 3077 + }, + { + "epoch": 0.370107617387122, + "grad_norm": 0.8761423691889074, + "learning_rate": 2.9045784107579123e-06, + "loss": 0.8473, + "step": 3078 + }, + { + "epoch": 0.37022786027776106, + "grad_norm": 1.8736293869440566, + "learning_rate": 2.9038835996820807e-06, + "loss": 0.9096, + "step": 3079 + }, + { + "epoch": 0.37034810316840017, + "grad_norm": 1.9067515354452906, + "learning_rate": 2.903188651487826e-06, + "loss": 1.0347, + "step": 3080 + }, + { + "epoch": 0.3704683460590393, + "grad_norm": 2.149676932968695, + "learning_rate": 2.902493566280571e-06, + "loss": 1.1047, + "step": 3081 + }, + { + "epoch": 0.37058858894967833, + "grad_norm": 1.9718614242255659, + "learning_rate": 2.9017983441657595e-06, + "loss": 1.0472, + "step": 3082 + }, + { + "epoch": 0.37070883184031744, + "grad_norm": 2.7440162060317435, + "learning_rate": 2.9011029852488564e-06, + "loss": 0.9891, + "step": 3083 + }, + { + "epoch": 0.37082907473095655, + "grad_norm": 0.9939530051307186, + "learning_rate": 2.9004074896353465e-06, + "loss": 0.862, + "step": 3084 + }, + { + "epoch": 0.3709493176215956, + "grad_norm": 3.639952503355582, + "learning_rate": 2.8997118574307362e-06, + "loss": 1.0559, + "step": 3085 + }, + { + "epoch": 0.3710695605122347, + "grad_norm": 2.6709844878734224, + "learning_rate": 2.899016088740553e-06, + "loss": 0.9838, + "step": 3086 + }, + { + "epoch": 0.37118980340287383, + "grad_norm": 2.015207291724448, + "learning_rate": 2.898320183670344e-06, + "loss": 1.0301, + "step": 3087 + }, + { + "epoch": 0.3713100462935129, + "grad_norm": 1.6849477328562081, + "learning_rate": 2.8976241423256767e-06, + "loss": 1.127, + "step": 3088 + }, + { + "epoch": 0.371430289184152, + "grad_norm": 2.105906368489116, + "learning_rate": 2.896927964812142e-06, + "loss": 0.9192, + "step": 3089 + }, + { + "epoch": 0.37155053207479105, + "grad_norm": 2.9595966775722133, + "learning_rate": 2.8962316512353465e-06, + "loss": 0.9926, + "step": 3090 + }, + { + "epoch": 0.37167077496543016, + "grad_norm": 1.8841692203244036, + "learning_rate": 2.8955352017009233e-06, + "loss": 0.9873, + "step": 3091 + }, + { + "epoch": 0.3717910178560693, + "grad_norm": 1.9448659606484238, + "learning_rate": 2.8948386163145212e-06, + "loss": 1.0098, + "step": 3092 + }, + { + "epoch": 0.3719112607467083, + "grad_norm": 2.111202550324363, + "learning_rate": 2.8941418951818135e-06, + "loss": 1.0317, + "step": 3093 + }, + { + "epoch": 0.37203150363734744, + "grad_norm": 2.2790147464571375, + "learning_rate": 2.8934450384084903e-06, + "loss": 0.9497, + "step": 3094 + }, + { + "epoch": 0.37215174652798655, + "grad_norm": 1.9538632637917555, + "learning_rate": 2.8927480461002653e-06, + "loss": 0.9334, + "step": 3095 + }, + { + "epoch": 0.3722719894186256, + "grad_norm": 2.752622394387131, + "learning_rate": 2.892050918362872e-06, + "loss": 1.0924, + "step": 3096 + }, + { + "epoch": 0.3723922323092647, + "grad_norm": 5.572419312527451, + "learning_rate": 2.8913536553020626e-06, + "loss": 0.7963, + "step": 3097 + }, + { + "epoch": 0.3725124751999038, + "grad_norm": 2.033786070819729, + "learning_rate": 2.8906562570236137e-06, + "loss": 1.0882, + "step": 3098 + }, + { + "epoch": 0.3726327180905429, + "grad_norm": 1.6789692607290045, + "learning_rate": 2.889958723633318e-06, + "loss": 1.0024, + "step": 3099 + }, + { + "epoch": 0.372752960981182, + "grad_norm": 1.8845369032204196, + "learning_rate": 2.889261055236992e-06, + "loss": 0.9773, + "step": 3100 + }, + { + "epoch": 0.3728732038718211, + "grad_norm": 1.7985028398378429, + "learning_rate": 2.8885632519404704e-06, + "loss": 1.065, + "step": 3101 + }, + { + "epoch": 0.37299344676246016, + "grad_norm": 2.2732096030198083, + "learning_rate": 2.8878653138496107e-06, + "loss": 0.9966, + "step": 3102 + }, + { + "epoch": 0.37311368965309927, + "grad_norm": 2.6015515026513056, + "learning_rate": 2.8871672410702878e-06, + "loss": 1.0018, + "step": 3103 + }, + { + "epoch": 0.3732339325437384, + "grad_norm": 1.7472383643278826, + "learning_rate": 2.8864690337084008e-06, + "loss": 1.0562, + "step": 3104 + }, + { + "epoch": 0.37335417543437743, + "grad_norm": 1.8616597219466837, + "learning_rate": 2.885770691869866e-06, + "loss": 1.0194, + "step": 3105 + }, + { + "epoch": 0.37347441832501654, + "grad_norm": 2.3379911753257065, + "learning_rate": 2.8850722156606207e-06, + "loss": 0.9854, + "step": 3106 + }, + { + "epoch": 0.3735946612156556, + "grad_norm": 2.017004561162358, + "learning_rate": 2.8843736051866252e-06, + "loss": 0.9057, + "step": 3107 + }, + { + "epoch": 0.3737149041062947, + "grad_norm": 1.805587147982437, + "learning_rate": 2.8836748605538557e-06, + "loss": 0.9273, + "step": 3108 + }, + { + "epoch": 0.3738351469969338, + "grad_norm": 2.1629708631918945, + "learning_rate": 2.882975981868313e-06, + "loss": 0.8715, + "step": 3109 + }, + { + "epoch": 0.3739553898875729, + "grad_norm": 2.1540728648833034, + "learning_rate": 2.882276969236016e-06, + "loss": 0.925, + "step": 3110 + }, + { + "epoch": 0.374075632778212, + "grad_norm": 2.0742751866365854, + "learning_rate": 2.881577822763005e-06, + "loss": 1.0008, + "step": 3111 + }, + { + "epoch": 0.3741958756688511, + "grad_norm": 1.8303018634702664, + "learning_rate": 2.880878542555338e-06, + "loss": 1.1125, + "step": 3112 + }, + { + "epoch": 0.37431611855949015, + "grad_norm": 3.181770411046089, + "learning_rate": 2.8801791287190976e-06, + "loss": 1.0402, + "step": 3113 + }, + { + "epoch": 0.37443636145012926, + "grad_norm": 2.701023144602697, + "learning_rate": 2.8794795813603817e-06, + "loss": 1.0923, + "step": 3114 + }, + { + "epoch": 0.3745566043407684, + "grad_norm": 1.7608969520429387, + "learning_rate": 2.878779900585314e-06, + "loss": 1.0563, + "step": 3115 + }, + { + "epoch": 0.37467684723140743, + "grad_norm": 1.59430064244418, + "learning_rate": 2.8780800865000336e-06, + "loss": 0.9875, + "step": 3116 + }, + { + "epoch": 0.37479709012204654, + "grad_norm": 0.9881658976787737, + "learning_rate": 2.877380139210702e-06, + "loss": 0.8295, + "step": 3117 + }, + { + "epoch": 0.37491733301268565, + "grad_norm": 1.7526040457451049, + "learning_rate": 2.876680058823501e-06, + "loss": 1.0003, + "step": 3118 + }, + { + "epoch": 0.3750375759033247, + "grad_norm": 1.8948726385793764, + "learning_rate": 2.8759798454446314e-06, + "loss": 0.8967, + "step": 3119 + }, + { + "epoch": 0.3751578187939638, + "grad_norm": 1.923575602577444, + "learning_rate": 2.8752794991803173e-06, + "loss": 1.0507, + "step": 3120 + }, + { + "epoch": 0.37527806168460287, + "grad_norm": 2.1133194953433314, + "learning_rate": 2.8745790201367976e-06, + "loss": 0.9908, + "step": 3121 + }, + { + "epoch": 0.375398304575242, + "grad_norm": 1.9241963455314575, + "learning_rate": 2.8738784084203373e-06, + "loss": 1.0785, + "step": 3122 + }, + { + "epoch": 0.3755185474658811, + "grad_norm": 3.421492185038252, + "learning_rate": 2.873177664137216e-06, + "loss": 1.0266, + "step": 3123 + }, + { + "epoch": 0.37563879035652015, + "grad_norm": 1.7924899789697657, + "learning_rate": 2.8724767873937384e-06, + "loss": 0.9292, + "step": 3124 + }, + { + "epoch": 0.37575903324715926, + "grad_norm": 2.6205120777244115, + "learning_rate": 2.871775778296225e-06, + "loss": 1.1116, + "step": 3125 + }, + { + "epoch": 0.37587927613779837, + "grad_norm": 2.7268378438435668, + "learning_rate": 2.8710746369510196e-06, + "loss": 1.0206, + "step": 3126 + }, + { + "epoch": 0.3759995190284374, + "grad_norm": 2.572766741092368, + "learning_rate": 2.8703733634644846e-06, + "loss": 1.0757, + "step": 3127 + }, + { + "epoch": 0.37611976191907653, + "grad_norm": 2.0734468737277516, + "learning_rate": 2.869671957943002e-06, + "loss": 1.0312, + "step": 3128 + }, + { + "epoch": 0.37624000480971564, + "grad_norm": 9.443280805322061, + "learning_rate": 2.8689704204929747e-06, + "loss": 0.9829, + "step": 3129 + }, + { + "epoch": 0.3763602477003547, + "grad_norm": 1.887264322889267, + "learning_rate": 2.8682687512208253e-06, + "loss": 1.0434, + "step": 3130 + }, + { + "epoch": 0.3764804905909938, + "grad_norm": 2.21869616376009, + "learning_rate": 2.8675669502329972e-06, + "loss": 1.0435, + "step": 3131 + }, + { + "epoch": 0.3766007334816329, + "grad_norm": 2.53483290292756, + "learning_rate": 2.866865017635952e-06, + "loss": 1.0874, + "step": 3132 + }, + { + "epoch": 0.376720976372272, + "grad_norm": 1.8735492179845965, + "learning_rate": 2.866162953536174e-06, + "loss": 1.0316, + "step": 3133 + }, + { + "epoch": 0.3768412192629111, + "grad_norm": 1.6132673581612849, + "learning_rate": 2.8654607580401634e-06, + "loss": 0.9861, + "step": 3134 + }, + { + "epoch": 0.3769614621535502, + "grad_norm": 0.8921822435855514, + "learning_rate": 2.8647584312544446e-06, + "loss": 0.8897, + "step": 3135 + }, + { + "epoch": 0.37708170504418925, + "grad_norm": 1.7217688664510056, + "learning_rate": 2.864055973285559e-06, + "loss": 1.0933, + "step": 3136 + }, + { + "epoch": 0.37720194793482836, + "grad_norm": 2.0456862164319958, + "learning_rate": 2.8633533842400698e-06, + "loss": 1.1034, + "step": 3137 + }, + { + "epoch": 0.3773221908254674, + "grad_norm": 2.016486168762641, + "learning_rate": 2.862650664224558e-06, + "loss": 1.0105, + "step": 3138 + }, + { + "epoch": 0.37744243371610653, + "grad_norm": 1.5076162576209238, + "learning_rate": 2.861947813345627e-06, + "loss": 0.9328, + "step": 3139 + }, + { + "epoch": 0.37756267660674564, + "grad_norm": 1.978688660837859, + "learning_rate": 2.8612448317098974e-06, + "loss": 0.9595, + "step": 3140 + }, + { + "epoch": 0.3776829194973847, + "grad_norm": 2.096045644736325, + "learning_rate": 2.8605417194240114e-06, + "loss": 1.0667, + "step": 3141 + }, + { + "epoch": 0.3778031623880238, + "grad_norm": 1.9804640549824157, + "learning_rate": 2.8598384765946315e-06, + "loss": 1.0266, + "step": 3142 + }, + { + "epoch": 0.3779234052786629, + "grad_norm": 1.977062056667804, + "learning_rate": 2.8591351033284377e-06, + "loss": 0.9553, + "step": 3143 + }, + { + "epoch": 0.37804364816930197, + "grad_norm": 2.36410956881121, + "learning_rate": 2.8584315997321325e-06, + "loss": 1.0731, + "step": 3144 + }, + { + "epoch": 0.3781638910599411, + "grad_norm": 2.506920676736594, + "learning_rate": 2.8577279659124356e-06, + "loss": 1.0205, + "step": 3145 + }, + { + "epoch": 0.3782841339505802, + "grad_norm": 2.624483057619665, + "learning_rate": 2.857024201976089e-06, + "loss": 1.0695, + "step": 3146 + }, + { + "epoch": 0.37840437684121925, + "grad_norm": 2.0541815189136727, + "learning_rate": 2.8563203080298516e-06, + "loss": 0.9675, + "step": 3147 + }, + { + "epoch": 0.37852461973185836, + "grad_norm": 3.0803741138693663, + "learning_rate": 2.855616284180505e-06, + "loss": 1.1356, + "step": 3148 + }, + { + "epoch": 0.37864486262249747, + "grad_norm": 0.8903831354294668, + "learning_rate": 2.8549121305348477e-06, + "loss": 0.9024, + "step": 3149 + }, + { + "epoch": 0.3787651055131365, + "grad_norm": 2.9870246987945204, + "learning_rate": 2.8542078471997006e-06, + "loss": 1.0709, + "step": 3150 + }, + { + "epoch": 0.37888534840377563, + "grad_norm": 1.7860306949946674, + "learning_rate": 2.8535034342819013e-06, + "loss": 0.9958, + "step": 3151 + }, + { + "epoch": 0.37900559129441475, + "grad_norm": 2.129077701489138, + "learning_rate": 2.85279889188831e-06, + "loss": 0.9664, + "step": 3152 + }, + { + "epoch": 0.3791258341850538, + "grad_norm": 2.2743316674775924, + "learning_rate": 2.852094220125805e-06, + "loss": 1.0475, + "step": 3153 + }, + { + "epoch": 0.3792460770756929, + "grad_norm": 3.016246830369064, + "learning_rate": 2.8513894191012846e-06, + "loss": 0.9485, + "step": 3154 + }, + { + "epoch": 0.37936631996633197, + "grad_norm": 1.5948984832505797, + "learning_rate": 2.8506844889216664e-06, + "loss": 1.0248, + "step": 3155 + }, + { + "epoch": 0.3794865628569711, + "grad_norm": 0.8703900218797523, + "learning_rate": 2.849979429693887e-06, + "loss": 0.8668, + "step": 3156 + }, + { + "epoch": 0.3796068057476102, + "grad_norm": 1.9452635977240018, + "learning_rate": 2.8492742415249042e-06, + "loss": 0.9838, + "step": 3157 + }, + { + "epoch": 0.37972704863824924, + "grad_norm": 1.7400057692815787, + "learning_rate": 2.848568924521694e-06, + "loss": 1.002, + "step": 3158 + }, + { + "epoch": 0.37984729152888835, + "grad_norm": 1.8977917487374445, + "learning_rate": 2.8478634787912526e-06, + "loss": 0.9732, + "step": 3159 + }, + { + "epoch": 0.37996753441952746, + "grad_norm": 2.2827292049180605, + "learning_rate": 2.847157904440596e-06, + "loss": 1.0003, + "step": 3160 + }, + { + "epoch": 0.3800877773101665, + "grad_norm": 1.9235236067288881, + "learning_rate": 2.846452201576759e-06, + "loss": 0.9745, + "step": 3161 + }, + { + "epoch": 0.38020802020080563, + "grad_norm": 0.9074085656650299, + "learning_rate": 2.845746370306795e-06, + "loss": 0.8656, + "step": 3162 + }, + { + "epoch": 0.38032826309144474, + "grad_norm": 1.9914420135507238, + "learning_rate": 2.84504041073778e-06, + "loss": 1.0191, + "step": 3163 + }, + { + "epoch": 0.3804485059820838, + "grad_norm": 1.9383337626479973, + "learning_rate": 2.844334322976806e-06, + "loss": 1.0305, + "step": 3164 + }, + { + "epoch": 0.3805687488727229, + "grad_norm": 1.8151605087431277, + "learning_rate": 2.8436281071309866e-06, + "loss": 1.0701, + "step": 3165 + }, + { + "epoch": 0.380688991763362, + "grad_norm": 0.7369186097911441, + "learning_rate": 2.842921763307455e-06, + "loss": 0.7673, + "step": 3166 + }, + { + "epoch": 0.38080923465400107, + "grad_norm": 2.059077573377466, + "learning_rate": 2.842215291613361e-06, + "loss": 1.0682, + "step": 3167 + }, + { + "epoch": 0.3809294775446402, + "grad_norm": 0.7780186775829014, + "learning_rate": 2.8415086921558774e-06, + "loss": 0.8299, + "step": 3168 + }, + { + "epoch": 0.38104972043527924, + "grad_norm": 1.5794398807050158, + "learning_rate": 2.840801965042194e-06, + "loss": 1.0255, + "step": 3169 + }, + { + "epoch": 0.38116996332591835, + "grad_norm": 1.944384740488098, + "learning_rate": 2.840095110379521e-06, + "loss": 1.0772, + "step": 3170 + }, + { + "epoch": 0.38129020621655746, + "grad_norm": 3.1658204269715107, + "learning_rate": 2.8393881282750884e-06, + "loss": 0.7764, + "step": 3171 + }, + { + "epoch": 0.3814104491071965, + "grad_norm": 2.570724036945078, + "learning_rate": 2.838681018836144e-06, + "loss": 1.0279, + "step": 3172 + }, + { + "epoch": 0.3815306919978356, + "grad_norm": 2.1233742048361206, + "learning_rate": 2.837973782169955e-06, + "loss": 1.0173, + "step": 3173 + }, + { + "epoch": 0.38165093488847474, + "grad_norm": 0.8157985195749686, + "learning_rate": 2.8372664183838096e-06, + "loss": 0.8294, + "step": 3174 + }, + { + "epoch": 0.3817711777791138, + "grad_norm": 2.249221349133669, + "learning_rate": 2.836558927585015e-06, + "loss": 0.9195, + "step": 3175 + }, + { + "epoch": 0.3818914206697529, + "grad_norm": 1.9480644278828396, + "learning_rate": 2.8358513098808957e-06, + "loss": 1.0618, + "step": 3176 + }, + { + "epoch": 0.382011663560392, + "grad_norm": 1.7574412992395585, + "learning_rate": 2.835143565378798e-06, + "loss": 1.006, + "step": 3177 + }, + { + "epoch": 0.38213190645103107, + "grad_norm": 2.377625469824013, + "learning_rate": 2.8344356941860847e-06, + "loss": 1.0238, + "step": 3178 + }, + { + "epoch": 0.3822521493416702, + "grad_norm": 2.0613260106865106, + "learning_rate": 2.8337276964101403e-06, + "loss": 0.8981, + "step": 3179 + }, + { + "epoch": 0.3823723922323093, + "grad_norm": 2.9245547485548844, + "learning_rate": 2.833019572158367e-06, + "loss": 1.0031, + "step": 3180 + }, + { + "epoch": 0.38249263512294834, + "grad_norm": 1.8343083072216575, + "learning_rate": 2.8323113215381872e-06, + "loss": 1.0386, + "step": 3181 + }, + { + "epoch": 0.38261287801358745, + "grad_norm": 2.4713799676596797, + "learning_rate": 2.831602944657042e-06, + "loss": 1.0007, + "step": 3182 + }, + { + "epoch": 0.38273312090422656, + "grad_norm": 2.203080846465797, + "learning_rate": 2.830894441622391e-06, + "loss": 0.9824, + "step": 3183 + }, + { + "epoch": 0.3828533637948656, + "grad_norm": 2.0075818831108783, + "learning_rate": 2.8301858125417134e-06, + "loss": 1.0353, + "step": 3184 + }, + { + "epoch": 0.38297360668550473, + "grad_norm": 2.1228473081729495, + "learning_rate": 2.8294770575225082e-06, + "loss": 0.9792, + "step": 3185 + }, + { + "epoch": 0.3830938495761438, + "grad_norm": 2.7373604307782933, + "learning_rate": 2.828768176672293e-06, + "loss": 1.0749, + "step": 3186 + }, + { + "epoch": 0.3832140924667829, + "grad_norm": 2.02289270556396, + "learning_rate": 2.8280591700986044e-06, + "loss": 0.9499, + "step": 3187 + }, + { + "epoch": 0.383334335357422, + "grad_norm": 1.7877511628298792, + "learning_rate": 2.827350037908999e-06, + "loss": 0.9857, + "step": 3188 + }, + { + "epoch": 0.38345457824806106, + "grad_norm": 2.1206443127810393, + "learning_rate": 2.8266407802110496e-06, + "loss": 1.029, + "step": 3189 + }, + { + "epoch": 0.3835748211387002, + "grad_norm": 2.5854362624350764, + "learning_rate": 2.8259313971123515e-06, + "loss": 0.9978, + "step": 3190 + }, + { + "epoch": 0.3836950640293393, + "grad_norm": 1.880551955787648, + "learning_rate": 2.8252218887205166e-06, + "loss": 1.0191, + "step": 3191 + }, + { + "epoch": 0.38381530691997834, + "grad_norm": 2.181138734794264, + "learning_rate": 2.824512255143178e-06, + "loss": 1.045, + "step": 3192 + }, + { + "epoch": 0.38393554981061745, + "grad_norm": 2.5223252056493823, + "learning_rate": 2.8238024964879855e-06, + "loss": 1.0311, + "step": 3193 + }, + { + "epoch": 0.38405579270125656, + "grad_norm": 2.3707770901284415, + "learning_rate": 2.8230926128626095e-06, + "loss": 1.0027, + "step": 3194 + }, + { + "epoch": 0.3841760355918956, + "grad_norm": 1.7988237511398373, + "learning_rate": 2.822382604374738e-06, + "loss": 1.0315, + "step": 3195 + }, + { + "epoch": 0.3842962784825347, + "grad_norm": 2.1361381118428255, + "learning_rate": 2.8216724711320793e-06, + "loss": 0.891, + "step": 3196 + }, + { + "epoch": 0.38441652137317384, + "grad_norm": 1.5976643084110647, + "learning_rate": 2.820962213242361e-06, + "loss": 1.0321, + "step": 3197 + }, + { + "epoch": 0.3845367642638129, + "grad_norm": 2.273613257676731, + "learning_rate": 2.8202518308133264e-06, + "loss": 1.0818, + "step": 3198 + }, + { + "epoch": 0.384657007154452, + "grad_norm": 1.8916524822134628, + "learning_rate": 2.8195413239527426e-06, + "loss": 0.9675, + "step": 3199 + }, + { + "epoch": 0.38477725004509106, + "grad_norm": 1.8476252805279791, + "learning_rate": 2.8188306927683906e-06, + "loss": 1.0428, + "step": 3200 + }, + { + "epoch": 0.38489749293573017, + "grad_norm": 2.2907389988446165, + "learning_rate": 2.818119937368074e-06, + "loss": 0.9853, + "step": 3201 + }, + { + "epoch": 0.3850177358263693, + "grad_norm": 1.9893568685641951, + "learning_rate": 2.817409057859613e-06, + "loss": 0.8982, + "step": 3202 + }, + { + "epoch": 0.38513797871700833, + "grad_norm": 1.8741204424029012, + "learning_rate": 2.8166980543508482e-06, + "loss": 1.0342, + "step": 3203 + }, + { + "epoch": 0.38525822160764744, + "grad_norm": 2.1324407791554787, + "learning_rate": 2.815986926949638e-06, + "loss": 1.0319, + "step": 3204 + }, + { + "epoch": 0.38537846449828655, + "grad_norm": 3.1111380787846223, + "learning_rate": 2.8152756757638597e-06, + "loss": 1.0393, + "step": 3205 + }, + { + "epoch": 0.3854987073889256, + "grad_norm": 1.9438687524367042, + "learning_rate": 2.8145643009014093e-06, + "loss": 1.0816, + "step": 3206 + }, + { + "epoch": 0.3856189502795647, + "grad_norm": 2.0451933963796605, + "learning_rate": 2.813852802470202e-06, + "loss": 1.0302, + "step": 3207 + }, + { + "epoch": 0.38573919317020383, + "grad_norm": 1.7341377852144624, + "learning_rate": 2.8131411805781717e-06, + "loss": 0.9554, + "step": 3208 + }, + { + "epoch": 0.3858594360608429, + "grad_norm": 2.6107002593396684, + "learning_rate": 2.8124294353332707e-06, + "loss": 0.88, + "step": 3209 + }, + { + "epoch": 0.385979678951482, + "grad_norm": 2.071194212673879, + "learning_rate": 2.8117175668434713e-06, + "loss": 1.0111, + "step": 3210 + }, + { + "epoch": 0.3860999218421211, + "grad_norm": 2.4861909958870387, + "learning_rate": 2.811005575216762e-06, + "loss": 0.939, + "step": 3211 + }, + { + "epoch": 0.38622016473276016, + "grad_norm": 1.4990119859090272, + "learning_rate": 2.8102934605611513e-06, + "loss": 1.0283, + "step": 3212 + }, + { + "epoch": 0.3863404076233993, + "grad_norm": 4.315082606014457, + "learning_rate": 2.8095812229846665e-06, + "loss": 0.9102, + "step": 3213 + }, + { + "epoch": 0.3864606505140384, + "grad_norm": 2.600731060979622, + "learning_rate": 2.808868862595355e-06, + "loss": 0.925, + "step": 3214 + }, + { + "epoch": 0.38658089340467744, + "grad_norm": 3.3412885786541584, + "learning_rate": 2.8081563795012795e-06, + "loss": 1.0346, + "step": 3215 + }, + { + "epoch": 0.38670113629531655, + "grad_norm": 2.0717394247907186, + "learning_rate": 2.807443773810524e-06, + "loss": 0.9774, + "step": 3216 + }, + { + "epoch": 0.3868213791859556, + "grad_norm": 1.9763314334467565, + "learning_rate": 2.80673104563119e-06, + "loss": 1.132, + "step": 3217 + }, + { + "epoch": 0.3869416220765947, + "grad_norm": 2.0288570687462464, + "learning_rate": 2.8060181950713976e-06, + "loss": 1.0236, + "step": 3218 + }, + { + "epoch": 0.3870618649672338, + "grad_norm": 2.8582216031523955, + "learning_rate": 2.805305222239286e-06, + "loss": 1.0485, + "step": 3219 + }, + { + "epoch": 0.3871821078578729, + "grad_norm": 2.0616904317133216, + "learning_rate": 2.8045921272430118e-06, + "loss": 0.9757, + "step": 3220 + }, + { + "epoch": 0.387302350748512, + "grad_norm": 2.039649265095171, + "learning_rate": 2.803878910190753e-06, + "loss": 1.0001, + "step": 3221 + }, + { + "epoch": 0.3874225936391511, + "grad_norm": 6.016358011347323, + "learning_rate": 2.8031655711907017e-06, + "loss": 1.0539, + "step": 3222 + }, + { + "epoch": 0.38754283652979016, + "grad_norm": 4.764963172858347, + "learning_rate": 2.8024521103510723e-06, + "loss": 1.0483, + "step": 3223 + }, + { + "epoch": 0.38766307942042927, + "grad_norm": 1.815725350684415, + "learning_rate": 2.8017385277800952e-06, + "loss": 0.9883, + "step": 3224 + }, + { + "epoch": 0.3877833223110684, + "grad_norm": 2.1662575506660953, + "learning_rate": 2.8010248235860213e-06, + "loss": 0.9867, + "step": 3225 + }, + { + "epoch": 0.38790356520170743, + "grad_norm": 0.8399664855697164, + "learning_rate": 2.8003109978771192e-06, + "loss": 0.8653, + "step": 3226 + }, + { + "epoch": 0.38802380809234654, + "grad_norm": 2.4396893376992987, + "learning_rate": 2.799597050761674e-06, + "loss": 1.0282, + "step": 3227 + }, + { + "epoch": 0.38814405098298566, + "grad_norm": 12.685933434497132, + "learning_rate": 2.7988829823479924e-06, + "loss": 1.0293, + "step": 3228 + }, + { + "epoch": 0.3882642938736247, + "grad_norm": 1.9785934688999, + "learning_rate": 2.7981687927443976e-06, + "loss": 0.8788, + "step": 3229 + }, + { + "epoch": 0.3883845367642638, + "grad_norm": 1.8626216464672, + "learning_rate": 2.797454482059231e-06, + "loss": 1.0988, + "step": 3230 + }, + { + "epoch": 0.3885047796549029, + "grad_norm": 1.626378135236586, + "learning_rate": 2.7967400504008537e-06, + "loss": 1.0822, + "step": 3231 + }, + { + "epoch": 0.388625022545542, + "grad_norm": 0.8332762380750034, + "learning_rate": 2.7960254978776456e-06, + "loss": 0.8126, + "step": 3232 + }, + { + "epoch": 0.3887452654361811, + "grad_norm": 2.2125891457394973, + "learning_rate": 2.7953108245980006e-06, + "loss": 1.0601, + "step": 3233 + }, + { + "epoch": 0.38886550832682015, + "grad_norm": 1.6789040091079377, + "learning_rate": 2.7945960306703365e-06, + "loss": 0.9755, + "step": 3234 + }, + { + "epoch": 0.38898575121745926, + "grad_norm": 1.753962390570391, + "learning_rate": 2.7938811162030865e-06, + "loss": 0.8959, + "step": 3235 + }, + { + "epoch": 0.3891059941080984, + "grad_norm": 2.4111818576674446, + "learning_rate": 2.793166081304702e-06, + "loss": 1.0596, + "step": 3236 + }, + { + "epoch": 0.38922623699873743, + "grad_norm": 1.9221834384935461, + "learning_rate": 2.7924509260836543e-06, + "loss": 1.0594, + "step": 3237 + }, + { + "epoch": 0.38934647988937654, + "grad_norm": 1.4652575099240386, + "learning_rate": 2.791735650648431e-06, + "loss": 0.9187, + "step": 3238 + }, + { + "epoch": 0.38946672278001565, + "grad_norm": 2.294773931401574, + "learning_rate": 2.791020255107538e-06, + "loss": 0.9836, + "step": 3239 + }, + { + "epoch": 0.3895869656706547, + "grad_norm": 1.715076891229186, + "learning_rate": 2.7903047395695023e-06, + "loss": 1.0457, + "step": 3240 + }, + { + "epoch": 0.3897072085612938, + "grad_norm": 2.1670910957528435, + "learning_rate": 2.789589104142865e-06, + "loss": 1.1406, + "step": 3241 + }, + { + "epoch": 0.3898274514519329, + "grad_norm": 1.8705850120626744, + "learning_rate": 2.7888733489361895e-06, + "loss": 1.0012, + "step": 3242 + }, + { + "epoch": 0.389947694342572, + "grad_norm": 0.8801814344159615, + "learning_rate": 2.788157474058054e-06, + "loss": 0.8252, + "step": 3243 + }, + { + "epoch": 0.3900679372332111, + "grad_norm": 1.5928806851203323, + "learning_rate": 2.7874414796170555e-06, + "loss": 0.9383, + "step": 3244 + }, + { + "epoch": 0.3901881801238502, + "grad_norm": 3.2838656218306514, + "learning_rate": 2.7867253657218113e-06, + "loss": 1.0761, + "step": 3245 + }, + { + "epoch": 0.39030842301448926, + "grad_norm": 7.132336770913212, + "learning_rate": 2.7860091324809544e-06, + "loss": 0.9643, + "step": 3246 + }, + { + "epoch": 0.39042866590512837, + "grad_norm": 1.9113763762183396, + "learning_rate": 2.7852927800031377e-06, + "loss": 1.0489, + "step": 3247 + }, + { + "epoch": 0.3905489087957674, + "grad_norm": 2.1582529467877754, + "learning_rate": 2.7845763083970298e-06, + "loss": 1.0673, + "step": 3248 + }, + { + "epoch": 0.39066915168640653, + "grad_norm": 1.8843224513380015, + "learning_rate": 2.7838597177713205e-06, + "loss": 1.0585, + "step": 3249 + }, + { + "epoch": 0.39078939457704565, + "grad_norm": 1.8247272183080216, + "learning_rate": 2.7831430082347143e-06, + "loss": 0.9741, + "step": 3250 + }, + { + "epoch": 0.3909096374676847, + "grad_norm": 2.045175560314998, + "learning_rate": 2.7824261798959373e-06, + "loss": 1.0616, + "step": 3251 + }, + { + "epoch": 0.3910298803583238, + "grad_norm": 2.049987988320493, + "learning_rate": 2.78170923286373e-06, + "loss": 1.0371, + "step": 3252 + }, + { + "epoch": 0.3911501232489629, + "grad_norm": 2.303722839195358, + "learning_rate": 2.780992167246854e-06, + "loss": 1.0758, + "step": 3253 + }, + { + "epoch": 0.391270366139602, + "grad_norm": 0.9847682288565555, + "learning_rate": 2.7802749831540883e-06, + "loss": 0.9596, + "step": 3254 + }, + { + "epoch": 0.3913906090302411, + "grad_norm": 2.068655596189702, + "learning_rate": 2.7795576806942268e-06, + "loss": 1.0581, + "step": 3255 + }, + { + "epoch": 0.3915108519208802, + "grad_norm": 0.8218471159755648, + "learning_rate": 2.778840259976085e-06, + "loss": 0.7871, + "step": 3256 + }, + { + "epoch": 0.39163109481151925, + "grad_norm": 2.1431850523056832, + "learning_rate": 2.778122721108495e-06, + "loss": 1.0001, + "step": 3257 + }, + { + "epoch": 0.39175133770215836, + "grad_norm": 2.1242470248698613, + "learning_rate": 2.7774050642003076e-06, + "loss": 1.121, + "step": 3258 + }, + { + "epoch": 0.3918715805927975, + "grad_norm": 2.262758006353884, + "learning_rate": 2.7766872893603896e-06, + "loss": 1.1694, + "step": 3259 + }, + { + "epoch": 0.39199182348343653, + "grad_norm": 1.6433270682679093, + "learning_rate": 2.7759693966976275e-06, + "loss": 0.9726, + "step": 3260 + }, + { + "epoch": 0.39211206637407564, + "grad_norm": 1.8435529183267017, + "learning_rate": 2.7752513863209242e-06, + "loss": 1.0931, + "step": 3261 + }, + { + "epoch": 0.39223230926471475, + "grad_norm": 1.6387844842676735, + "learning_rate": 2.774533258339203e-06, + "loss": 1.0815, + "step": 3262 + }, + { + "epoch": 0.3923525521553538, + "grad_norm": 2.3589222454664016, + "learning_rate": 2.7738150128614014e-06, + "loss": 1.0305, + "step": 3263 + }, + { + "epoch": 0.3924727950459929, + "grad_norm": 2.293425064562589, + "learning_rate": 2.7730966499964777e-06, + "loss": 1.1351, + "step": 3264 + }, + { + "epoch": 0.39259303793663197, + "grad_norm": 2.6352847120298506, + "learning_rate": 2.772378169853408e-06, + "loss": 1.0498, + "step": 3265 + }, + { + "epoch": 0.3927132808272711, + "grad_norm": 2.0787398497149585, + "learning_rate": 2.771659572541183e-06, + "loss": 0.9804, + "step": 3266 + }, + { + "epoch": 0.3928335237179102, + "grad_norm": 1.9640057029796705, + "learning_rate": 2.7709408581688143e-06, + "loss": 1.1071, + "step": 3267 + }, + { + "epoch": 0.39295376660854925, + "grad_norm": 1.6886869424134063, + "learning_rate": 2.7702220268453307e-06, + "loss": 1.1166, + "step": 3268 + }, + { + "epoch": 0.39307400949918836, + "grad_norm": 2.787825451656069, + "learning_rate": 2.7695030786797785e-06, + "loss": 1.0845, + "step": 3269 + }, + { + "epoch": 0.39319425238982747, + "grad_norm": 1.9703699138362507, + "learning_rate": 2.7687840137812206e-06, + "loss": 0.9829, + "step": 3270 + }, + { + "epoch": 0.3933144952804665, + "grad_norm": 0.8087982803626415, + "learning_rate": 2.7680648322587395e-06, + "loss": 0.8575, + "step": 3271 + }, + { + "epoch": 0.39343473817110564, + "grad_norm": 1.9619396430209202, + "learning_rate": 2.7673455342214334e-06, + "loss": 1.0461, + "step": 3272 + }, + { + "epoch": 0.39355498106174475, + "grad_norm": 2.7111088151696214, + "learning_rate": 2.7666261197784198e-06, + "loss": 0.9986, + "step": 3273 + }, + { + "epoch": 0.3936752239523838, + "grad_norm": 2.065391825155328, + "learning_rate": 2.7659065890388336e-06, + "loss": 1.0008, + "step": 3274 + }, + { + "epoch": 0.3937954668430229, + "grad_norm": 3.4262700531409926, + "learning_rate": 2.7651869421118266e-06, + "loss": 1.0847, + "step": 3275 + }, + { + "epoch": 0.393915709733662, + "grad_norm": 3.0458452098387014, + "learning_rate": 2.76446717910657e-06, + "loss": 1.0677, + "step": 3276 + }, + { + "epoch": 0.3940359526243011, + "grad_norm": 3.322801083714033, + "learning_rate": 2.763747300132249e-06, + "loss": 1.0118, + "step": 3277 + }, + { + "epoch": 0.3941561955149402, + "grad_norm": 2.7239518660887367, + "learning_rate": 2.7630273052980704e-06, + "loss": 1.1008, + "step": 3278 + }, + { + "epoch": 0.39427643840557924, + "grad_norm": 2.1546430266440018, + "learning_rate": 2.7623071947132554e-06, + "loss": 0.9101, + "step": 3279 + }, + { + "epoch": 0.39439668129621835, + "grad_norm": 2.1757069545916106, + "learning_rate": 2.7615869684870458e-06, + "loss": 1.0264, + "step": 3280 + }, + { + "epoch": 0.39451692418685746, + "grad_norm": 1.7037352422353556, + "learning_rate": 2.7608666267286986e-06, + "loss": 1.0854, + "step": 3281 + }, + { + "epoch": 0.3946371670774965, + "grad_norm": 2.8803150968901035, + "learning_rate": 2.760146169547489e-06, + "loss": 1.1035, + "step": 3282 + }, + { + "epoch": 0.39475740996813563, + "grad_norm": 1.6210482303708915, + "learning_rate": 2.75942559705271e-06, + "loss": 1.0052, + "step": 3283 + }, + { + "epoch": 0.39487765285877474, + "grad_norm": 1.8723095983207776, + "learning_rate": 2.7587049093536713e-06, + "loss": 1.1287, + "step": 3284 + }, + { + "epoch": 0.3949978957494138, + "grad_norm": 2.009065467166534, + "learning_rate": 2.757984106559701e-06, + "loss": 1.0467, + "step": 3285 + }, + { + "epoch": 0.3951181386400529, + "grad_norm": 2.5576499071682854, + "learning_rate": 2.7572631887801446e-06, + "loss": 0.9471, + "step": 3286 + }, + { + "epoch": 0.395238381530692, + "grad_norm": 2.0910556594832976, + "learning_rate": 2.7565421561243654e-06, + "loss": 1.0037, + "step": 3287 + }, + { + "epoch": 0.3953586244213311, + "grad_norm": 2.1359213504027252, + "learning_rate": 2.7558210087017413e-06, + "loss": 1.0559, + "step": 3288 + }, + { + "epoch": 0.3954788673119702, + "grad_norm": 2.0870171085491735, + "learning_rate": 2.7550997466216724e-06, + "loss": 0.9727, + "step": 3289 + }, + { + "epoch": 0.3955991102026093, + "grad_norm": 2.2033695599246483, + "learning_rate": 2.7543783699935714e-06, + "loss": 1.0514, + "step": 3290 + }, + { + "epoch": 0.39571935309324835, + "grad_norm": 2.536212530634867, + "learning_rate": 2.753656878926872e-06, + "loss": 1.0964, + "step": 3291 + }, + { + "epoch": 0.39583959598388746, + "grad_norm": 1.8468589433465845, + "learning_rate": 2.752935273531023e-06, + "loss": 0.9791, + "step": 3292 + }, + { + "epoch": 0.39595983887452657, + "grad_norm": 1.7573830991368837, + "learning_rate": 2.752213553915492e-06, + "loss": 1.024, + "step": 3293 + }, + { + "epoch": 0.3960800817651656, + "grad_norm": 0.8297274578120721, + "learning_rate": 2.751491720189762e-06, + "loss": 0.8986, + "step": 3294 + }, + { + "epoch": 0.39620032465580474, + "grad_norm": 2.1549944331900437, + "learning_rate": 2.7507697724633364e-06, + "loss": 1.154, + "step": 3295 + }, + { + "epoch": 0.3963205675464438, + "grad_norm": 0.7845093068798, + "learning_rate": 2.7500477108457327e-06, + "loss": 0.7848, + "step": 3296 + }, + { + "epoch": 0.3964408104370829, + "grad_norm": 1.9025612458084298, + "learning_rate": 2.7493255354464877e-06, + "loss": 1.0428, + "step": 3297 + }, + { + "epoch": 0.396561053327722, + "grad_norm": 1.774062731260547, + "learning_rate": 2.748603246375156e-06, + "loss": 0.9977, + "step": 3298 + }, + { + "epoch": 0.39668129621836107, + "grad_norm": 2.1799261584593754, + "learning_rate": 2.7478808437413055e-06, + "loss": 0.9323, + "step": 3299 + }, + { + "epoch": 0.3968015391090002, + "grad_norm": 1.9249689273060357, + "learning_rate": 2.7471583276545263e-06, + "loss": 0.8996, + "step": 3300 + }, + { + "epoch": 0.3969217819996393, + "grad_norm": 2.1135472607965515, + "learning_rate": 2.7464356982244224e-06, + "loss": 0.9424, + "step": 3301 + }, + { + "epoch": 0.39704202489027834, + "grad_norm": 0.7849794634787228, + "learning_rate": 2.745712955560617e-06, + "loss": 0.8556, + "step": 3302 + }, + { + "epoch": 0.39716226778091746, + "grad_norm": 2.5515170241050904, + "learning_rate": 2.7449900997727496e-06, + "loss": 1.0124, + "step": 3303 + }, + { + "epoch": 0.39728251067155657, + "grad_norm": 1.7502864194540053, + "learning_rate": 2.744267130970476e-06, + "loss": 1.0794, + "step": 3304 + }, + { + "epoch": 0.3974027535621956, + "grad_norm": 2.097581460709921, + "learning_rate": 2.7435440492634697e-06, + "loss": 1.0085, + "step": 3305 + }, + { + "epoch": 0.39752299645283473, + "grad_norm": 2.3181035994187016, + "learning_rate": 2.7428208547614228e-06, + "loss": 0.9105, + "step": 3306 + }, + { + "epoch": 0.39764323934347384, + "grad_norm": 2.611850865953852, + "learning_rate": 2.742097547574043e-06, + "loss": 1.0081, + "step": 3307 + }, + { + "epoch": 0.3977634822341129, + "grad_norm": 1.827296729622474, + "learning_rate": 2.7413741278110544e-06, + "loss": 1.0158, + "step": 3308 + }, + { + "epoch": 0.397883725124752, + "grad_norm": 2.1429513138559937, + "learning_rate": 2.7406505955822016e-06, + "loss": 0.9258, + "step": 3309 + }, + { + "epoch": 0.39800396801539106, + "grad_norm": 2.3754707672947637, + "learning_rate": 2.7399269509972415e-06, + "loss": 0.9001, + "step": 3310 + }, + { + "epoch": 0.3981242109060302, + "grad_norm": 2.718642484797154, + "learning_rate": 2.7392031941659514e-06, + "loss": 1.086, + "step": 3311 + }, + { + "epoch": 0.3982444537966693, + "grad_norm": 1.9438798523732495, + "learning_rate": 2.7384793251981244e-06, + "loss": 1.0973, + "step": 3312 + }, + { + "epoch": 0.39836469668730834, + "grad_norm": 2.0027832691942127, + "learning_rate": 2.737755344203571e-06, + "loss": 1.0469, + "step": 3313 + }, + { + "epoch": 0.39848493957794745, + "grad_norm": 1.8500383115358423, + "learning_rate": 2.7370312512921186e-06, + "loss": 1.0333, + "step": 3314 + }, + { + "epoch": 0.39860518246858656, + "grad_norm": 2.566225553564292, + "learning_rate": 2.736307046573611e-06, + "loss": 1.0011, + "step": 3315 + }, + { + "epoch": 0.3987254253592256, + "grad_norm": 1.686576166526975, + "learning_rate": 2.73558273015791e-06, + "loss": 1.0587, + "step": 3316 + }, + { + "epoch": 0.3988456682498647, + "grad_norm": 4.2769972569419155, + "learning_rate": 2.734858302154894e-06, + "loss": 0.9413, + "step": 3317 + }, + { + "epoch": 0.39896591114050384, + "grad_norm": 2.097542814743045, + "learning_rate": 2.734133762674457e-06, + "loss": 0.9994, + "step": 3318 + }, + { + "epoch": 0.3990861540311429, + "grad_norm": 2.3809337776456396, + "learning_rate": 2.7334091118265124e-06, + "loss": 0.9399, + "step": 3319 + }, + { + "epoch": 0.399206396921782, + "grad_norm": 0.6889891751951899, + "learning_rate": 2.732684349720989e-06, + "loss": 0.8167, + "step": 3320 + }, + { + "epoch": 0.3993266398124211, + "grad_norm": 1.799295404070949, + "learning_rate": 2.7319594764678318e-06, + "loss": 0.9951, + "step": 3321 + }, + { + "epoch": 0.39944688270306017, + "grad_norm": 2.3463025244400706, + "learning_rate": 2.7312344921770044e-06, + "loss": 1.0702, + "step": 3322 + }, + { + "epoch": 0.3995671255936993, + "grad_norm": 1.9349834957946805, + "learning_rate": 2.7305093969584857e-06, + "loss": 1.0242, + "step": 3323 + }, + { + "epoch": 0.3996873684843384, + "grad_norm": 1.6774944804480263, + "learning_rate": 2.729784190922272e-06, + "loss": 1.0342, + "step": 3324 + }, + { + "epoch": 0.39980761137497745, + "grad_norm": 1.2703056907705919, + "learning_rate": 2.729058874178378e-06, + "loss": 0.8109, + "step": 3325 + }, + { + "epoch": 0.39992785426561656, + "grad_norm": 1.9371521545228034, + "learning_rate": 2.7283334468368315e-06, + "loss": 0.9317, + "step": 3326 + }, + { + "epoch": 0.4000480971562556, + "grad_norm": 1.9444692231673328, + "learning_rate": 2.72760790900768e-06, + "loss": 0.9662, + "step": 3327 + }, + { + "epoch": 0.4001683400468947, + "grad_norm": 1.7953744344760212, + "learning_rate": 2.7268822608009875e-06, + "loss": 1.0261, + "step": 3328 + }, + { + "epoch": 0.40028858293753383, + "grad_norm": 2.0706975614785663, + "learning_rate": 2.726156502326834e-06, + "loss": 1.0188, + "step": 3329 + }, + { + "epoch": 0.4004088258281729, + "grad_norm": 0.7032571668385129, + "learning_rate": 2.725430633695316e-06, + "loss": 0.8404, + "step": 3330 + }, + { + "epoch": 0.400529068718812, + "grad_norm": 0.912663718129177, + "learning_rate": 2.7247046550165485e-06, + "loss": 0.8176, + "step": 3331 + }, + { + "epoch": 0.4006493116094511, + "grad_norm": 1.5857149917031463, + "learning_rate": 2.7239785664006606e-06, + "loss": 0.9974, + "step": 3332 + }, + { + "epoch": 0.40076955450009016, + "grad_norm": 0.7764110240916171, + "learning_rate": 2.7232523679578002e-06, + "loss": 0.8557, + "step": 3333 + }, + { + "epoch": 0.4008897973907293, + "grad_norm": 2.5899354391999103, + "learning_rate": 2.7225260597981295e-06, + "loss": 1.0314, + "step": 3334 + }, + { + "epoch": 0.4010100402813684, + "grad_norm": 2.925432130893797, + "learning_rate": 2.721799642031831e-06, + "loss": 1.0255, + "step": 3335 + }, + { + "epoch": 0.40113028317200744, + "grad_norm": 3.206871409736665, + "learning_rate": 2.721073114769101e-06, + "loss": 1.0168, + "step": 3336 + }, + { + "epoch": 0.40125052606264655, + "grad_norm": 1.9454123842877595, + "learning_rate": 2.7203464781201523e-06, + "loss": 0.9891, + "step": 3337 + }, + { + "epoch": 0.40137076895328566, + "grad_norm": 2.9899793798954044, + "learning_rate": 2.719619732195215e-06, + "loss": 1.0205, + "step": 3338 + }, + { + "epoch": 0.4014910118439247, + "grad_norm": 1.5084216216717417, + "learning_rate": 2.7188928771045377e-06, + "loss": 0.9661, + "step": 3339 + }, + { + "epoch": 0.4016112547345638, + "grad_norm": 1.5796815353265854, + "learning_rate": 2.7181659129583815e-06, + "loss": 1.0373, + "step": 3340 + }, + { + "epoch": 0.4017314976252029, + "grad_norm": 1.9565858010386796, + "learning_rate": 2.7174388398670276e-06, + "loss": 0.9979, + "step": 3341 + }, + { + "epoch": 0.401851740515842, + "grad_norm": 2.036803181386712, + "learning_rate": 2.716711657940773e-06, + "loss": 1.1568, + "step": 3342 + }, + { + "epoch": 0.4019719834064811, + "grad_norm": 0.8206134788942105, + "learning_rate": 2.7159843672899284e-06, + "loss": 0.8029, + "step": 3343 + }, + { + "epoch": 0.40209222629712016, + "grad_norm": 1.8979215195471997, + "learning_rate": 2.715256968024825e-06, + "loss": 1.05, + "step": 3344 + }, + { + "epoch": 0.40221246918775927, + "grad_norm": 1.5933819514136653, + "learning_rate": 2.7145294602558083e-06, + "loss": 1.0628, + "step": 3345 + }, + { + "epoch": 0.4023327120783984, + "grad_norm": 2.5716939535648486, + "learning_rate": 2.713801844093241e-06, + "loss": 0.9406, + "step": 3346 + }, + { + "epoch": 0.40245295496903744, + "grad_norm": 2.0795909328954942, + "learning_rate": 2.7130741196475014e-06, + "loss": 1.1254, + "step": 3347 + }, + { + "epoch": 0.40257319785967655, + "grad_norm": 1.8291974304786784, + "learning_rate": 2.7123462870289848e-06, + "loss": 1.0313, + "step": 3348 + }, + { + "epoch": 0.40269344075031566, + "grad_norm": 2.1265683837629847, + "learning_rate": 2.711618346348102e-06, + "loss": 1.0497, + "step": 3349 + }, + { + "epoch": 0.4028136836409547, + "grad_norm": 2.3869167314443795, + "learning_rate": 2.7108902977152825e-06, + "loss": 0.874, + "step": 3350 + }, + { + "epoch": 0.4029339265315938, + "grad_norm": 2.4192501487436235, + "learning_rate": 2.7101621412409704e-06, + "loss": 0.9851, + "step": 3351 + }, + { + "epoch": 0.40305416942223293, + "grad_norm": 2.5179221522664808, + "learning_rate": 2.7094338770356256e-06, + "loss": 1.0909, + "step": 3352 + }, + { + "epoch": 0.403174412312872, + "grad_norm": 9.382151406508447, + "learning_rate": 2.708705505209726e-06, + "loss": 0.8791, + "step": 3353 + }, + { + "epoch": 0.4032946552035111, + "grad_norm": 2.208711615737863, + "learning_rate": 2.7079770258737646e-06, + "loss": 1.1546, + "step": 3354 + }, + { + "epoch": 0.4034148980941502, + "grad_norm": 2.065624067389989, + "learning_rate": 2.707248439138251e-06, + "loss": 0.9904, + "step": 3355 + }, + { + "epoch": 0.40353514098478926, + "grad_norm": 5.920660614064413, + "learning_rate": 2.7065197451137114e-06, + "loss": 0.89, + "step": 3356 + }, + { + "epoch": 0.4036553838754284, + "grad_norm": 2.1126043395509493, + "learning_rate": 2.7057909439106894e-06, + "loss": 0.913, + "step": 3357 + }, + { + "epoch": 0.40377562676606743, + "grad_norm": 1.977980335122415, + "learning_rate": 2.7050620356397417e-06, + "loss": 1.0226, + "step": 3358 + }, + { + "epoch": 0.40389586965670654, + "grad_norm": 1.7831977100436336, + "learning_rate": 2.7043330204114437e-06, + "loss": 0.9626, + "step": 3359 + }, + { + "epoch": 0.40401611254734565, + "grad_norm": 2.1406091842623995, + "learning_rate": 2.7036038983363862e-06, + "loss": 1.0984, + "step": 3360 + }, + { + "epoch": 0.4041363554379847, + "grad_norm": 1.7184024112529992, + "learning_rate": 2.702874669525177e-06, + "loss": 1.0792, + "step": 3361 + }, + { + "epoch": 0.4042565983286238, + "grad_norm": 1.9753779906701832, + "learning_rate": 2.7021453340884394e-06, + "loss": 0.9335, + "step": 3362 + }, + { + "epoch": 0.40437684121926293, + "grad_norm": 2.4806721281042603, + "learning_rate": 2.7014158921368125e-06, + "loss": 0.9672, + "step": 3363 + }, + { + "epoch": 0.404497084109902, + "grad_norm": 1.8510760863507791, + "learning_rate": 2.700686343780953e-06, + "loss": 1.0916, + "step": 3364 + }, + { + "epoch": 0.4046173270005411, + "grad_norm": 1.7332381913165122, + "learning_rate": 2.699956689131532e-06, + "loss": 1.1168, + "step": 3365 + }, + { + "epoch": 0.4047375698911802, + "grad_norm": 2.9428713773047193, + "learning_rate": 2.699226928299238e-06, + "loss": 1.0882, + "step": 3366 + }, + { + "epoch": 0.40485781278181926, + "grad_norm": 2.5174483805178167, + "learning_rate": 2.698497061394774e-06, + "loss": 1.0204, + "step": 3367 + }, + { + "epoch": 0.40497805567245837, + "grad_norm": 1.7377876180902454, + "learning_rate": 2.6977670885288627e-06, + "loss": 1.0452, + "step": 3368 + }, + { + "epoch": 0.4050982985630975, + "grad_norm": 2.0994977486678517, + "learning_rate": 2.6970370098122378e-06, + "loss": 0.9933, + "step": 3369 + }, + { + "epoch": 0.40521854145373654, + "grad_norm": 1.623471421446356, + "learning_rate": 2.6963068253556535e-06, + "loss": 1.1061, + "step": 3370 + }, + { + "epoch": 0.40533878434437565, + "grad_norm": 1.9607619856750549, + "learning_rate": 2.6955765352698763e-06, + "loss": 1.0942, + "step": 3371 + }, + { + "epoch": 0.40545902723501476, + "grad_norm": 4.209160424464936, + "learning_rate": 2.6948461396656923e-06, + "loss": 0.968, + "step": 3372 + }, + { + "epoch": 0.4055792701256538, + "grad_norm": 2.4729936077314503, + "learning_rate": 2.6941156386539013e-06, + "loss": 0.9819, + "step": 3373 + }, + { + "epoch": 0.4056995130162929, + "grad_norm": 2.9408746368139473, + "learning_rate": 2.6933850323453203e-06, + "loss": 1.0424, + "step": 3374 + }, + { + "epoch": 0.405819755906932, + "grad_norm": 1.8023731535194336, + "learning_rate": 2.6926543208507806e-06, + "loss": 0.9826, + "step": 3375 + }, + { + "epoch": 0.4059399987975711, + "grad_norm": 2.283861311782105, + "learning_rate": 2.6919235042811316e-06, + "loss": 1.0334, + "step": 3376 + }, + { + "epoch": 0.4060602416882102, + "grad_norm": 2.0759141677723596, + "learning_rate": 2.691192582747237e-06, + "loss": 1.0001, + "step": 3377 + }, + { + "epoch": 0.40618048457884925, + "grad_norm": 1.8191957772978438, + "learning_rate": 2.6904615563599765e-06, + "loss": 0.9749, + "step": 3378 + }, + { + "epoch": 0.40630072746948837, + "grad_norm": 3.1629229718844276, + "learning_rate": 2.6897304252302477e-06, + "loss": 1.0699, + "step": 3379 + }, + { + "epoch": 0.4064209703601275, + "grad_norm": 0.7955784441084238, + "learning_rate": 2.688999189468962e-06, + "loss": 0.7857, + "step": 3380 + }, + { + "epoch": 0.40654121325076653, + "grad_norm": 3.6775251056532516, + "learning_rate": 2.6882678491870464e-06, + "loss": 1.0007, + "step": 3381 + }, + { + "epoch": 0.40666145614140564, + "grad_norm": 1.7252125705826193, + "learning_rate": 2.6875364044954453e-06, + "loss": 0.9505, + "step": 3382 + }, + { + "epoch": 0.40678169903204475, + "grad_norm": 3.6203671809756326, + "learning_rate": 2.6868048555051185e-06, + "loss": 1.0576, + "step": 3383 + }, + { + "epoch": 0.4069019419226838, + "grad_norm": 2.5981936709111007, + "learning_rate": 2.686073202327041e-06, + "loss": 1.0923, + "step": 3384 + }, + { + "epoch": 0.4070221848133229, + "grad_norm": 1.7565469212529194, + "learning_rate": 2.6853414450722043e-06, + "loss": 0.9694, + "step": 3385 + }, + { + "epoch": 0.40714242770396203, + "grad_norm": 2.099252445133662, + "learning_rate": 2.684609583851616e-06, + "loss": 1.0882, + "step": 3386 + }, + { + "epoch": 0.4072626705946011, + "grad_norm": 1.8582000867118318, + "learning_rate": 2.683877618776297e-06, + "loss": 1.0423, + "step": 3387 + }, + { + "epoch": 0.4073829134852402, + "grad_norm": 2.8710680071656296, + "learning_rate": 2.6831455499572876e-06, + "loss": 0.9807, + "step": 3388 + }, + { + "epoch": 0.40750315637587925, + "grad_norm": 1.906620114776329, + "learning_rate": 2.682413377505641e-06, + "loss": 1.0146, + "step": 3389 + }, + { + "epoch": 0.40762339926651836, + "grad_norm": 1.8060701696508463, + "learning_rate": 2.6816811015324284e-06, + "loss": 1.004, + "step": 3390 + }, + { + "epoch": 0.40774364215715747, + "grad_norm": 0.7360009394878113, + "learning_rate": 2.6809487221487343e-06, + "loss": 0.8049, + "step": 3391 + }, + { + "epoch": 0.4078638850477965, + "grad_norm": 2.308691858572589, + "learning_rate": 2.6802162394656605e-06, + "loss": 1.0527, + "step": 3392 + }, + { + "epoch": 0.40798412793843564, + "grad_norm": 1.7173259037184874, + "learning_rate": 2.679483653594324e-06, + "loss": 0.9559, + "step": 3393 + }, + { + "epoch": 0.40810437082907475, + "grad_norm": 2.348471907657383, + "learning_rate": 2.678750964645857e-06, + "loss": 0.9999, + "step": 3394 + }, + { + "epoch": 0.4082246137197138, + "grad_norm": 2.553303101396326, + "learning_rate": 2.6780181727314094e-06, + "loss": 1.0754, + "step": 3395 + }, + { + "epoch": 0.4083448566103529, + "grad_norm": 1.8818568420040902, + "learning_rate": 2.6772852779621435e-06, + "loss": 1.015, + "step": 3396 + }, + { + "epoch": 0.408465099500992, + "grad_norm": 2.215800729516263, + "learning_rate": 2.676552280449239e-06, + "loss": 1.104, + "step": 3397 + }, + { + "epoch": 0.4085853423916311, + "grad_norm": 4.218377669116618, + "learning_rate": 2.6758191803038917e-06, + "loss": 0.9925, + "step": 3398 + }, + { + "epoch": 0.4087055852822702, + "grad_norm": 2.2200938708417537, + "learning_rate": 2.6750859776373125e-06, + "loss": 1.0641, + "step": 3399 + }, + { + "epoch": 0.4088258281729093, + "grad_norm": 0.777687916692487, + "learning_rate": 2.674352672560727e-06, + "loss": 0.8414, + "step": 3400 + }, + { + "epoch": 0.40894607106354836, + "grad_norm": 1.7794392400038155, + "learning_rate": 2.673619265185377e-06, + "loss": 1.0042, + "step": 3401 + }, + { + "epoch": 0.40906631395418747, + "grad_norm": 1.6984363376925788, + "learning_rate": 2.672885755622521e-06, + "loss": 1.0184, + "step": 3402 + }, + { + "epoch": 0.4091865568448266, + "grad_norm": 7.369001114750847, + "learning_rate": 2.67215214398343e-06, + "loss": 0.9414, + "step": 3403 + }, + { + "epoch": 0.40930679973546563, + "grad_norm": 2.275550830096481, + "learning_rate": 2.671418430379393e-06, + "loss": 1.0202, + "step": 3404 + }, + { + "epoch": 0.40942704262610474, + "grad_norm": 1.9089911716590755, + "learning_rate": 2.670684614921715e-06, + "loss": 1.0694, + "step": 3405 + }, + { + "epoch": 0.4095472855167438, + "grad_norm": 2.571665306395861, + "learning_rate": 2.6699506977217128e-06, + "loss": 0.9332, + "step": 3406 + }, + { + "epoch": 0.4096675284073829, + "grad_norm": 2.3258759561002567, + "learning_rate": 2.6692166788907233e-06, + "loss": 0.9397, + "step": 3407 + }, + { + "epoch": 0.409787771298022, + "grad_norm": 1.8707787362042536, + "learning_rate": 2.6684825585400957e-06, + "loss": 1.0058, + "step": 3408 + }, + { + "epoch": 0.4099080141886611, + "grad_norm": 0.8219923711594431, + "learning_rate": 2.6677483367811947e-06, + "loss": 0.8892, + "step": 3409 + }, + { + "epoch": 0.4100282570793002, + "grad_norm": 2.1886416044904413, + "learning_rate": 2.6670140137254028e-06, + "loss": 0.9942, + "step": 3410 + }, + { + "epoch": 0.4101484999699393, + "grad_norm": 2.373249010809774, + "learning_rate": 2.666279589484115e-06, + "loss": 1.1325, + "step": 3411 + }, + { + "epoch": 0.41026874286057835, + "grad_norm": 1.9464820501997644, + "learning_rate": 2.6655450641687435e-06, + "loss": 1.0472, + "step": 3412 + }, + { + "epoch": 0.41038898575121746, + "grad_norm": 1.7884143171552582, + "learning_rate": 2.664810437890715e-06, + "loss": 0.9338, + "step": 3413 + }, + { + "epoch": 0.41050922864185657, + "grad_norm": 1.8955157366768292, + "learning_rate": 2.6640757107614714e-06, + "loss": 1.0314, + "step": 3414 + }, + { + "epoch": 0.4106294715324956, + "grad_norm": 2.4880380556563124, + "learning_rate": 2.6633408828924697e-06, + "loss": 0.9313, + "step": 3415 + }, + { + "epoch": 0.41074971442313474, + "grad_norm": 1.5649107074553326, + "learning_rate": 2.662605954395185e-06, + "loss": 0.9375, + "step": 3416 + }, + { + "epoch": 0.41086995731377385, + "grad_norm": 1.7426837847208692, + "learning_rate": 2.6618709253811027e-06, + "loss": 1.0714, + "step": 3417 + }, + { + "epoch": 0.4109902002044129, + "grad_norm": 1.650677501416943, + "learning_rate": 2.6611357959617277e-06, + "loss": 1.1131, + "step": 3418 + }, + { + "epoch": 0.411110443095052, + "grad_norm": 1.8865427128586794, + "learning_rate": 2.660400566248578e-06, + "loss": 1.148, + "step": 3419 + }, + { + "epoch": 0.41123068598569107, + "grad_norm": 2.4794840142084174, + "learning_rate": 2.6596652363531876e-06, + "loss": 0.9102, + "step": 3420 + }, + { + "epoch": 0.4113509288763302, + "grad_norm": 1.5351121425358907, + "learning_rate": 2.6589298063871055e-06, + "loss": 1.0241, + "step": 3421 + }, + { + "epoch": 0.4114711717669693, + "grad_norm": 1.899042731617377, + "learning_rate": 2.658194276461895e-06, + "loss": 0.9301, + "step": 3422 + }, + { + "epoch": 0.41159141465760835, + "grad_norm": 2.042475040801567, + "learning_rate": 2.6574586466891368e-06, + "loss": 0.9102, + "step": 3423 + }, + { + "epoch": 0.41171165754824746, + "grad_norm": 2.1829714084418197, + "learning_rate": 2.6567229171804247e-06, + "loss": 0.8844, + "step": 3424 + }, + { + "epoch": 0.41183190043888657, + "grad_norm": 3.042413820097813, + "learning_rate": 2.655987088047368e-06, + "loss": 1.1145, + "step": 3425 + }, + { + "epoch": 0.4119521433295256, + "grad_norm": 2.2797497558459754, + "learning_rate": 2.6552511594015912e-06, + "loss": 1.0237, + "step": 3426 + }, + { + "epoch": 0.41207238622016473, + "grad_norm": 1.9616807235866656, + "learning_rate": 2.654515131354735e-06, + "loss": 1.0879, + "step": 3427 + }, + { + "epoch": 0.41219262911080384, + "grad_norm": 2.3956498338390966, + "learning_rate": 2.653779004018453e-06, + "loss": 1.0847, + "step": 3428 + }, + { + "epoch": 0.4123128720014429, + "grad_norm": 2.6151936238014812, + "learning_rate": 2.653042777504417e-06, + "loss": 1.0599, + "step": 3429 + }, + { + "epoch": 0.412433114892082, + "grad_norm": 2.770954148567445, + "learning_rate": 2.6523064519243105e-06, + "loss": 1.0328, + "step": 3430 + }, + { + "epoch": 0.4125533577827211, + "grad_norm": 2.4518037384358196, + "learning_rate": 2.6515700273898333e-06, + "loss": 1.0309, + "step": 3431 + }, + { + "epoch": 0.4126736006733602, + "grad_norm": 2.4696710534304156, + "learning_rate": 2.6508335040127018e-06, + "loss": 0.9297, + "step": 3432 + }, + { + "epoch": 0.4127938435639993, + "grad_norm": 1.613265806405916, + "learning_rate": 2.6500968819046446e-06, + "loss": 1.0079, + "step": 3433 + }, + { + "epoch": 0.4129140864546384, + "grad_norm": 2.460250807126392, + "learning_rate": 2.649360161177408e-06, + "loss": 0.832, + "step": 3434 + }, + { + "epoch": 0.41303432934527745, + "grad_norm": 1.7478148578127954, + "learning_rate": 2.6486233419427504e-06, + "loss": 0.9711, + "step": 3435 + }, + { + "epoch": 0.41315457223591656, + "grad_norm": 2.11992195597533, + "learning_rate": 2.6478864243124484e-06, + "loss": 0.9928, + "step": 3436 + }, + { + "epoch": 0.4132748151265556, + "grad_norm": 2.459414637632753, + "learning_rate": 2.6471494083982903e-06, + "loss": 1.0882, + "step": 3437 + }, + { + "epoch": 0.4133950580171947, + "grad_norm": 1.9673204833054534, + "learning_rate": 2.6464122943120818e-06, + "loss": 0.9888, + "step": 3438 + }, + { + "epoch": 0.41351530090783384, + "grad_norm": 3.965059261384256, + "learning_rate": 2.645675082165642e-06, + "loss": 1.0557, + "step": 3439 + }, + { + "epoch": 0.4136355437984729, + "grad_norm": 2.6091503115100476, + "learning_rate": 2.644937772070806e-06, + "loss": 0.9933, + "step": 3440 + }, + { + "epoch": 0.413755786689112, + "grad_norm": 2.301959579616672, + "learning_rate": 2.6442003641394225e-06, + "loss": 1.0697, + "step": 3441 + }, + { + "epoch": 0.4138760295797511, + "grad_norm": 2.9039783418546596, + "learning_rate": 2.643462858483356e-06, + "loss": 1.0767, + "step": 3442 + }, + { + "epoch": 0.41399627247039017, + "grad_norm": 1.967486142372664, + "learning_rate": 2.6427252552144856e-06, + "loss": 0.9672, + "step": 3443 + }, + { + "epoch": 0.4141165153610293, + "grad_norm": 2.033828624478665, + "learning_rate": 2.6419875544447044e-06, + "loss": 0.9828, + "step": 3444 + }, + { + "epoch": 0.4142367582516684, + "grad_norm": 2.247117845678264, + "learning_rate": 2.6412497562859218e-06, + "loss": 0.9575, + "step": 3445 + }, + { + "epoch": 0.41435700114230745, + "grad_norm": 2.4756699165982745, + "learning_rate": 2.6405118608500617e-06, + "loss": 0.9982, + "step": 3446 + }, + { + "epoch": 0.41447724403294656, + "grad_norm": 1.7538040913011559, + "learning_rate": 2.6397738682490613e-06, + "loss": 1.0502, + "step": 3447 + }, + { + "epoch": 0.41459748692358567, + "grad_norm": 1.8622300483617056, + "learning_rate": 2.6390357785948734e-06, + "loss": 0.9912, + "step": 3448 + }, + { + "epoch": 0.4147177298142247, + "grad_norm": 1.938255285898689, + "learning_rate": 2.6382975919994667e-06, + "loss": 1.0435, + "step": 3449 + }, + { + "epoch": 0.41483797270486383, + "grad_norm": 1.7290945689172925, + "learning_rate": 2.637559308574822e-06, + "loss": 0.9658, + "step": 3450 + }, + { + "epoch": 0.4149582155955029, + "grad_norm": 2.0238611668855024, + "learning_rate": 2.6368209284329376e-06, + "loss": 0.9493, + "step": 3451 + }, + { + "epoch": 0.415078458486142, + "grad_norm": 2.036226953758926, + "learning_rate": 2.636082451685825e-06, + "loss": 0.9938, + "step": 3452 + }, + { + "epoch": 0.4151987013767811, + "grad_norm": 1.8898345303635822, + "learning_rate": 2.6353438784455094e-06, + "loss": 1.1, + "step": 3453 + }, + { + "epoch": 0.41531894426742016, + "grad_norm": 2.477038990476201, + "learning_rate": 2.6346052088240326e-06, + "loss": 0.9549, + "step": 3454 + }, + { + "epoch": 0.4154391871580593, + "grad_norm": 2.026482136235162, + "learning_rate": 2.63386644293345e-06, + "loss": 1.0088, + "step": 3455 + }, + { + "epoch": 0.4155594300486984, + "grad_norm": 2.765497840501173, + "learning_rate": 2.633127580885833e-06, + "loss": 1.0693, + "step": 3456 + }, + { + "epoch": 0.41567967293933744, + "grad_norm": 2.1391222057271895, + "learning_rate": 2.632388622793265e-06, + "loss": 0.8882, + "step": 3457 + }, + { + "epoch": 0.41579991582997655, + "grad_norm": 1.712009111758967, + "learning_rate": 2.6316495687678457e-06, + "loss": 0.9122, + "step": 3458 + }, + { + "epoch": 0.41592015872061566, + "grad_norm": 2.482649677908217, + "learning_rate": 2.6309104189216887e-06, + "loss": 1.0078, + "step": 3459 + }, + { + "epoch": 0.4160404016112547, + "grad_norm": 2.3068119925361694, + "learning_rate": 2.630171173366923e-06, + "loss": 0.9848, + "step": 3460 + }, + { + "epoch": 0.41616064450189383, + "grad_norm": 2.707682037292356, + "learning_rate": 2.629431832215691e-06, + "loss": 0.984, + "step": 3461 + }, + { + "epoch": 0.41628088739253294, + "grad_norm": 2.928880259881102, + "learning_rate": 2.628692395580151e-06, + "loss": 1.1083, + "step": 3462 + }, + { + "epoch": 0.416401130283172, + "grad_norm": 2.0971308499516303, + "learning_rate": 2.6279528635724747e-06, + "loss": 1.0284, + "step": 3463 + }, + { + "epoch": 0.4165213731738111, + "grad_norm": 3.241151207456568, + "learning_rate": 2.627213236304848e-06, + "loss": 1.0235, + "step": 3464 + }, + { + "epoch": 0.4166416160644502, + "grad_norm": 2.430802068168158, + "learning_rate": 2.626473513889472e-06, + "loss": 0.946, + "step": 3465 + }, + { + "epoch": 0.41676185895508927, + "grad_norm": 2.4401845576106362, + "learning_rate": 2.625733696438562e-06, + "loss": 1.0667, + "step": 3466 + }, + { + "epoch": 0.4168821018457284, + "grad_norm": 1.7276186927397092, + "learning_rate": 2.6249937840643476e-06, + "loss": 0.9891, + "step": 3467 + }, + { + "epoch": 0.41700234473636744, + "grad_norm": 1.830831757262496, + "learning_rate": 2.6242537768790733e-06, + "loss": 0.9038, + "step": 3468 + }, + { + "epoch": 0.41712258762700655, + "grad_norm": 1.9027535312465949, + "learning_rate": 2.6235136749949975e-06, + "loss": 0.9232, + "step": 3469 + }, + { + "epoch": 0.41724283051764566, + "grad_norm": 2.0885717781414, + "learning_rate": 2.6227734785243924e-06, + "loss": 0.8518, + "step": 3470 + }, + { + "epoch": 0.4173630734082847, + "grad_norm": 1.8339200887087226, + "learning_rate": 2.6220331875795466e-06, + "loss": 1.0261, + "step": 3471 + }, + { + "epoch": 0.4174833162989238, + "grad_norm": 1.6981012988972501, + "learning_rate": 2.62129280227276e-06, + "loss": 0.9884, + "step": 3472 + }, + { + "epoch": 0.41760355918956293, + "grad_norm": 1.941577128636589, + "learning_rate": 2.62055232271635e-06, + "loss": 0.9218, + "step": 3473 + }, + { + "epoch": 0.417723802080202, + "grad_norm": 2.366099897763467, + "learning_rate": 2.619811749022646e-06, + "loss": 1.1161, + "step": 3474 + }, + { + "epoch": 0.4178440449708411, + "grad_norm": 2.1700737159267547, + "learning_rate": 2.6190710813039917e-06, + "loss": 0.9517, + "step": 3475 + }, + { + "epoch": 0.4179642878614802, + "grad_norm": 2.8790523331534166, + "learning_rate": 2.618330319672747e-06, + "loss": 1.0728, + "step": 3476 + }, + { + "epoch": 0.41808453075211927, + "grad_norm": 2.4972868502123764, + "learning_rate": 2.617589464241284e-06, + "loss": 1.1576, + "step": 3477 + }, + { + "epoch": 0.4182047736427584, + "grad_norm": 2.0246753809195703, + "learning_rate": 2.6168485151219914e-06, + "loss": 0.9842, + "step": 3478 + }, + { + "epoch": 0.4183250165333975, + "grad_norm": 2.321574577518873, + "learning_rate": 2.616107472427269e-06, + "loss": 0.9537, + "step": 3479 + }, + { + "epoch": 0.41844525942403654, + "grad_norm": 2.596651591435533, + "learning_rate": 2.615366336269533e-06, + "loss": 1.0067, + "step": 3480 + }, + { + "epoch": 0.41856550231467565, + "grad_norm": 2.4141527671356773, + "learning_rate": 2.6146251067612126e-06, + "loss": 1.0487, + "step": 3481 + }, + { + "epoch": 0.41868574520531476, + "grad_norm": 1.6235265654143012, + "learning_rate": 2.6138837840147525e-06, + "loss": 1.0637, + "step": 3482 + }, + { + "epoch": 0.4188059880959538, + "grad_norm": 2.0758603024603586, + "learning_rate": 2.6131423681426103e-06, + "loss": 1.0046, + "step": 3483 + }, + { + "epoch": 0.41892623098659293, + "grad_norm": 1.9721219689524287, + "learning_rate": 2.6124008592572587e-06, + "loss": 0.9666, + "step": 3484 + }, + { + "epoch": 0.419046473877232, + "grad_norm": 2.1644660215435927, + "learning_rate": 2.6116592574711835e-06, + "loss": 1.0576, + "step": 3485 + }, + { + "epoch": 0.4191667167678711, + "grad_norm": 1.8445086453970163, + "learning_rate": 2.6109175628968853e-06, + "loss": 1.077, + "step": 3486 + }, + { + "epoch": 0.4192869596585102, + "grad_norm": 1.981552038425322, + "learning_rate": 2.610175775646878e-06, + "loss": 1.0699, + "step": 3487 + }, + { + "epoch": 0.41940720254914926, + "grad_norm": 1.9987856039057463, + "learning_rate": 2.6094338958336907e-06, + "loss": 0.967, + "step": 3488 + }, + { + "epoch": 0.41952744543978837, + "grad_norm": 2.0560383000721374, + "learning_rate": 2.608691923569867e-06, + "loss": 1.0627, + "step": 3489 + }, + { + "epoch": 0.4196476883304275, + "grad_norm": 1.7223902213627422, + "learning_rate": 2.6079498589679616e-06, + "loss": 0.998, + "step": 3490 + }, + { + "epoch": 0.41976793122106654, + "grad_norm": 1.8931016070084585, + "learning_rate": 2.6072077021405465e-06, + "loss": 0.9955, + "step": 3491 + }, + { + "epoch": 0.41988817411170565, + "grad_norm": 1.8026281058915132, + "learning_rate": 2.6064654532002054e-06, + "loss": 0.9347, + "step": 3492 + }, + { + "epoch": 0.42000841700234476, + "grad_norm": 1.5172948388768333, + "learning_rate": 2.6057231122595375e-06, + "loss": 0.9992, + "step": 3493 + }, + { + "epoch": 0.4201286598929838, + "grad_norm": 1.9072266875079964, + "learning_rate": 2.604980679431154e-06, + "loss": 0.9632, + "step": 3494 + }, + { + "epoch": 0.4202489027836229, + "grad_norm": 2.2490685972316875, + "learning_rate": 2.604238154827684e-06, + "loss": 0.981, + "step": 3495 + }, + { + "epoch": 0.42036914567426203, + "grad_norm": 2.087314908165126, + "learning_rate": 2.6034955385617656e-06, + "loss": 0.9658, + "step": 3496 + }, + { + "epoch": 0.4204893885649011, + "grad_norm": 0.7277130942885592, + "learning_rate": 2.6027528307460544e-06, + "loss": 0.8542, + "step": 3497 + }, + { + "epoch": 0.4206096314555402, + "grad_norm": 1.926073342353905, + "learning_rate": 2.602010031493217e-06, + "loss": 1.1029, + "step": 3498 + }, + { + "epoch": 0.42072987434617926, + "grad_norm": 2.6220564914672986, + "learning_rate": 2.6012671409159367e-06, + "loss": 1.1069, + "step": 3499 + }, + { + "epoch": 0.42085011723681837, + "grad_norm": 1.9140895919439507, + "learning_rate": 2.6005241591269097e-06, + "loss": 1.0543, + "step": 3500 + }, + { + "epoch": 0.4209703601274575, + "grad_norm": 2.0751942784923894, + "learning_rate": 2.5997810862388454e-06, + "loss": 1.033, + "step": 3501 + }, + { + "epoch": 0.42109060301809653, + "grad_norm": 2.1157113309795257, + "learning_rate": 2.599037922364467e-06, + "loss": 0.9965, + "step": 3502 + }, + { + "epoch": 0.42121084590873564, + "grad_norm": 2.2129338286341347, + "learning_rate": 2.5982946676165112e-06, + "loss": 0.9878, + "step": 3503 + }, + { + "epoch": 0.42133108879937475, + "grad_norm": 0.7350318081778523, + "learning_rate": 2.5975513221077313e-06, + "loss": 0.8144, + "step": 3504 + }, + { + "epoch": 0.4214513316900138, + "grad_norm": 2.4301056823983007, + "learning_rate": 2.5968078859508897e-06, + "loss": 1.1251, + "step": 3505 + }, + { + "epoch": 0.4215715745806529, + "grad_norm": 1.9549273300323462, + "learning_rate": 2.5960643592587673e-06, + "loss": 1.0343, + "step": 3506 + }, + { + "epoch": 0.42169181747129203, + "grad_norm": 1.9208408457622237, + "learning_rate": 2.5953207421441553e-06, + "loss": 1.0529, + "step": 3507 + }, + { + "epoch": 0.4218120603619311, + "grad_norm": 2.5162742203495725, + "learning_rate": 2.5945770347198603e-06, + "loss": 0.9902, + "step": 3508 + }, + { + "epoch": 0.4219323032525702, + "grad_norm": 1.932436386680054, + "learning_rate": 2.593833237098701e-06, + "loss": 1.0544, + "step": 3509 + }, + { + "epoch": 0.4220525461432093, + "grad_norm": 1.95507477141711, + "learning_rate": 2.593089349393512e-06, + "loss": 0.8636, + "step": 3510 + }, + { + "epoch": 0.42217278903384836, + "grad_norm": 2.0100588570702898, + "learning_rate": 2.592345371717141e-06, + "loss": 1.0716, + "step": 3511 + }, + { + "epoch": 0.42229303192448747, + "grad_norm": 2.1625737877754, + "learning_rate": 2.591601304182448e-06, + "loss": 0.9601, + "step": 3512 + }, + { + "epoch": 0.4224132748151266, + "grad_norm": 1.8068347393605502, + "learning_rate": 2.5908571469023067e-06, + "loss": 1.0294, + "step": 3513 + }, + { + "epoch": 0.42253351770576564, + "grad_norm": 2.261817416079742, + "learning_rate": 2.5901128999896067e-06, + "loss": 0.987, + "step": 3514 + }, + { + "epoch": 0.42265376059640475, + "grad_norm": 1.7065873954785826, + "learning_rate": 2.5893685635572487e-06, + "loss": 0.92, + "step": 3515 + }, + { + "epoch": 0.4227740034870438, + "grad_norm": 1.9529479900257471, + "learning_rate": 2.5886241377181483e-06, + "loss": 0.9305, + "step": 3516 + }, + { + "epoch": 0.4228942463776829, + "grad_norm": 1.7807291172768056, + "learning_rate": 2.587879622585234e-06, + "loss": 1.0526, + "step": 3517 + }, + { + "epoch": 0.423014489268322, + "grad_norm": 2.0416150248737, + "learning_rate": 2.5871350182714486e-06, + "loss": 0.9985, + "step": 3518 + }, + { + "epoch": 0.4231347321589611, + "grad_norm": 2.17092578371356, + "learning_rate": 2.586390324889748e-06, + "loss": 1.0405, + "step": 3519 + }, + { + "epoch": 0.4232549750496002, + "grad_norm": 2.1099588222011056, + "learning_rate": 2.5856455425531003e-06, + "loss": 0.9144, + "step": 3520 + }, + { + "epoch": 0.4233752179402393, + "grad_norm": 1.8798065249334175, + "learning_rate": 2.5849006713744902e-06, + "loss": 1.0448, + "step": 3521 + }, + { + "epoch": 0.42349546083087836, + "grad_norm": 2.280739400497606, + "learning_rate": 2.5841557114669135e-06, + "loss": 0.9628, + "step": 3522 + }, + { + "epoch": 0.42361570372151747, + "grad_norm": 2.8851718913526394, + "learning_rate": 2.58341066294338e-06, + "loss": 0.9054, + "step": 3523 + }, + { + "epoch": 0.4237359466121566, + "grad_norm": 2.2054593320152924, + "learning_rate": 2.5826655259169124e-06, + "loss": 1.0893, + "step": 3524 + }, + { + "epoch": 0.42385618950279563, + "grad_norm": 2.4046310328942173, + "learning_rate": 2.5819203005005475e-06, + "loss": 1.1445, + "step": 3525 + }, + { + "epoch": 0.42397643239343474, + "grad_norm": 1.6314883178610684, + "learning_rate": 2.581174986807336e-06, + "loss": 1.0246, + "step": 3526 + }, + { + "epoch": 0.42409667528407385, + "grad_norm": 2.579903468906252, + "learning_rate": 2.580429584950341e-06, + "loss": 1.1497, + "step": 3527 + }, + { + "epoch": 0.4242169181747129, + "grad_norm": 2.446101539718813, + "learning_rate": 2.5796840950426397e-06, + "loss": 0.8996, + "step": 3528 + }, + { + "epoch": 0.424337161065352, + "grad_norm": 1.903248158746508, + "learning_rate": 2.578938517197322e-06, + "loss": 0.8912, + "step": 3529 + }, + { + "epoch": 0.4244574039559911, + "grad_norm": 2.1607528331790538, + "learning_rate": 2.5781928515274916e-06, + "loss": 0.8578, + "step": 3530 + }, + { + "epoch": 0.4245776468466302, + "grad_norm": 1.9070899863693453, + "learning_rate": 2.577447098146265e-06, + "loss": 0.9174, + "step": 3531 + }, + { + "epoch": 0.4246978897372693, + "grad_norm": 1.8066384475282486, + "learning_rate": 2.5767012571667724e-06, + "loss": 1.0289, + "step": 3532 + }, + { + "epoch": 0.42481813262790835, + "grad_norm": 1.8142967526300662, + "learning_rate": 2.5759553287021587e-06, + "loss": 0.9194, + "step": 3533 + }, + { + "epoch": 0.42493837551854746, + "grad_norm": 2.0571191127859985, + "learning_rate": 2.5752093128655786e-06, + "loss": 1.015, + "step": 3534 + }, + { + "epoch": 0.4250586184091866, + "grad_norm": 1.9369830395532552, + "learning_rate": 2.574463209770204e-06, + "loss": 0.9714, + "step": 3535 + }, + { + "epoch": 0.42517886129982563, + "grad_norm": 1.891544024095057, + "learning_rate": 2.5737170195292165e-06, + "loss": 1.0363, + "step": 3536 + }, + { + "epoch": 0.42529910419046474, + "grad_norm": 1.7918024529153422, + "learning_rate": 2.572970742255814e-06, + "loss": 1.0187, + "step": 3537 + }, + { + "epoch": 0.42541934708110385, + "grad_norm": 1.9032512278420703, + "learning_rate": 2.5722243780632046e-06, + "loss": 1.0505, + "step": 3538 + }, + { + "epoch": 0.4255395899717429, + "grad_norm": 0.7560732832060275, + "learning_rate": 2.5714779270646125e-06, + "loss": 0.8422, + "step": 3539 + }, + { + "epoch": 0.425659832862382, + "grad_norm": 2.461226479653827, + "learning_rate": 2.5707313893732735e-06, + "loss": 1.0141, + "step": 3540 + }, + { + "epoch": 0.4257800757530211, + "grad_norm": 2.3284778711579306, + "learning_rate": 2.5699847651024364e-06, + "loss": 1.0097, + "step": 3541 + }, + { + "epoch": 0.4259003186436602, + "grad_norm": 2.1633277499573085, + "learning_rate": 2.5692380543653627e-06, + "loss": 1.0047, + "step": 3542 + }, + { + "epoch": 0.4260205615342993, + "grad_norm": 2.2223207994955096, + "learning_rate": 2.5684912572753293e-06, + "loss": 0.9374, + "step": 3543 + }, + { + "epoch": 0.4261408044249384, + "grad_norm": 1.911603320978205, + "learning_rate": 2.5677443739456245e-06, + "loss": 1.0786, + "step": 3544 + }, + { + "epoch": 0.42626104731557746, + "grad_norm": 2.255911795927726, + "learning_rate": 2.5669974044895495e-06, + "loss": 1.0304, + "step": 3545 + }, + { + "epoch": 0.42638129020621657, + "grad_norm": 1.7701796129119198, + "learning_rate": 2.5662503490204187e-06, + "loss": 1.0302, + "step": 3546 + }, + { + "epoch": 0.4265015330968556, + "grad_norm": 2.0219631641628504, + "learning_rate": 2.5655032076515603e-06, + "loss": 0.9998, + "step": 3547 + }, + { + "epoch": 0.42662177598749473, + "grad_norm": 2.0698037728075738, + "learning_rate": 2.5647559804963155e-06, + "loss": 1.056, + "step": 3548 + }, + { + "epoch": 0.42674201887813384, + "grad_norm": 2.558125739937458, + "learning_rate": 2.5640086676680364e-06, + "loss": 1.0261, + "step": 3549 + }, + { + "epoch": 0.4268622617687729, + "grad_norm": 2.0521210281343856, + "learning_rate": 2.5632612692800923e-06, + "loss": 1.04, + "step": 3550 + }, + { + "epoch": 0.426982504659412, + "grad_norm": 2.5361809445037946, + "learning_rate": 2.5625137854458603e-06, + "loss": 0.9899, + "step": 3551 + }, + { + "epoch": 0.4271027475500511, + "grad_norm": 1.9749549087978358, + "learning_rate": 2.561766216278735e-06, + "loss": 1.0375, + "step": 3552 + }, + { + "epoch": 0.4272229904406902, + "grad_norm": 2.165881992961316, + "learning_rate": 2.561018561892121e-06, + "loss": 1.0511, + "step": 3553 + }, + { + "epoch": 0.4273432333313293, + "grad_norm": 1.7080908657533902, + "learning_rate": 2.5602708223994363e-06, + "loss": 1.0061, + "step": 3554 + }, + { + "epoch": 0.4274634762219684, + "grad_norm": 2.8964224748948784, + "learning_rate": 2.559522997914115e-06, + "loss": 0.9141, + "step": 3555 + }, + { + "epoch": 0.42758371911260745, + "grad_norm": 2.2981954917854868, + "learning_rate": 2.558775088549599e-06, + "loss": 1.0819, + "step": 3556 + }, + { + "epoch": 0.42770396200324656, + "grad_norm": 2.886148316387606, + "learning_rate": 2.5580270944193467e-06, + "loss": 0.9034, + "step": 3557 + }, + { + "epoch": 0.4278242048938857, + "grad_norm": 0.7573093434981327, + "learning_rate": 2.557279015636827e-06, + "loss": 0.7928, + "step": 3558 + }, + { + "epoch": 0.42794444778452473, + "grad_norm": 0.7776710549620576, + "learning_rate": 2.5565308523155245e-06, + "loss": 0.8504, + "step": 3559 + }, + { + "epoch": 0.42806469067516384, + "grad_norm": 3.285883946510668, + "learning_rate": 2.5557826045689336e-06, + "loss": 1.0586, + "step": 3560 + }, + { + "epoch": 0.4281849335658029, + "grad_norm": 0.8307591415944656, + "learning_rate": 2.5550342725105643e-06, + "loss": 0.8259, + "step": 3561 + }, + { + "epoch": 0.428305176456442, + "grad_norm": 1.7041310829731737, + "learning_rate": 2.554285856253937e-06, + "loss": 1.049, + "step": 3562 + }, + { + "epoch": 0.4284254193470811, + "grad_norm": 2.0055937029285884, + "learning_rate": 2.5535373559125855e-06, + "loss": 1.0103, + "step": 3563 + }, + { + "epoch": 0.42854566223772017, + "grad_norm": 1.8654476982479786, + "learning_rate": 2.552788771600057e-06, + "loss": 1.0538, + "step": 3564 + }, + { + "epoch": 0.4286659051283593, + "grad_norm": 2.0273867916764092, + "learning_rate": 2.5520401034299118e-06, + "loss": 1.0604, + "step": 3565 + }, + { + "epoch": 0.4287861480189984, + "grad_norm": 2.0742463363278554, + "learning_rate": 2.551291351515722e-06, + "loss": 1.1154, + "step": 3566 + }, + { + "epoch": 0.42890639090963745, + "grad_norm": 1.755480631035604, + "learning_rate": 2.5505425159710726e-06, + "loss": 1.0939, + "step": 3567 + }, + { + "epoch": 0.42902663380027656, + "grad_norm": 2.5811355122075925, + "learning_rate": 2.549793596909561e-06, + "loss": 1.062, + "step": 3568 + }, + { + "epoch": 0.42914687669091567, + "grad_norm": 2.0161574495835466, + "learning_rate": 2.5490445944447976e-06, + "loss": 0.9006, + "step": 3569 + }, + { + "epoch": 0.4292671195815547, + "grad_norm": 2.0880101796848427, + "learning_rate": 2.548295508690406e-06, + "loss": 0.8868, + "step": 3570 + }, + { + "epoch": 0.42938736247219383, + "grad_norm": 1.8467848418509543, + "learning_rate": 2.5475463397600217e-06, + "loss": 1.0007, + "step": 3571 + }, + { + "epoch": 0.42950760536283294, + "grad_norm": 2.2579587473247518, + "learning_rate": 2.546797087767293e-06, + "loss": 1.0143, + "step": 3572 + }, + { + "epoch": 0.429627848253472, + "grad_norm": 1.995157339445455, + "learning_rate": 2.546047752825881e-06, + "loss": 1.1117, + "step": 3573 + }, + { + "epoch": 0.4297480911441111, + "grad_norm": 2.1948322028671523, + "learning_rate": 2.5452983350494595e-06, + "loss": 1.1675, + "step": 3574 + }, + { + "epoch": 0.4298683340347502, + "grad_norm": 2.0903141811350974, + "learning_rate": 2.544548834551713e-06, + "loss": 0.8838, + "step": 3575 + }, + { + "epoch": 0.4299885769253893, + "grad_norm": 2.278978340230121, + "learning_rate": 2.5437992514463424e-06, + "loss": 1.1821, + "step": 3576 + }, + { + "epoch": 0.4301088198160284, + "grad_norm": 2.768237243430029, + "learning_rate": 2.5430495858470565e-06, + "loss": 1.1169, + "step": 3577 + }, + { + "epoch": 0.43022906270666744, + "grad_norm": 3.021011680512259, + "learning_rate": 2.54229983786758e-06, + "loss": 1.0102, + "step": 3578 + }, + { + "epoch": 0.43034930559730655, + "grad_norm": 2.339645555664652, + "learning_rate": 2.541550007621651e-06, + "loss": 1.0878, + "step": 3579 + }, + { + "epoch": 0.43046954848794566, + "grad_norm": 1.9702541115781673, + "learning_rate": 2.5408000952230156e-06, + "loss": 1.0395, + "step": 3580 + }, + { + "epoch": 0.4305897913785847, + "grad_norm": 2.180457303964662, + "learning_rate": 2.5400501007854357e-06, + "loss": 1.139, + "step": 3581 + }, + { + "epoch": 0.43071003426922383, + "grad_norm": 1.9079810240699617, + "learning_rate": 2.539300024422685e-06, + "loss": 0.9915, + "step": 3582 + }, + { + "epoch": 0.43083027715986294, + "grad_norm": 0.801199008550997, + "learning_rate": 2.538549866248549e-06, + "loss": 0.8472, + "step": 3583 + }, + { + "epoch": 0.430950520050502, + "grad_norm": 1.897379827802741, + "learning_rate": 2.5377996263768274e-06, + "loss": 1.0463, + "step": 3584 + }, + { + "epoch": 0.4310707629411411, + "grad_norm": 1.774801662270793, + "learning_rate": 2.5370493049213293e-06, + "loss": 0.9232, + "step": 3585 + }, + { + "epoch": 0.4311910058317802, + "grad_norm": 2.2891674863936275, + "learning_rate": 2.536298901995878e-06, + "loss": 1.0365, + "step": 3586 + }, + { + "epoch": 0.43131124872241927, + "grad_norm": 1.7525343255179142, + "learning_rate": 2.535548417714311e-06, + "loss": 1.0397, + "step": 3587 + }, + { + "epoch": 0.4314314916130584, + "grad_norm": 1.5992052548225328, + "learning_rate": 2.534797852190474e-06, + "loss": 1.107, + "step": 3588 + }, + { + "epoch": 0.4315517345036975, + "grad_norm": 2.275550096678312, + "learning_rate": 2.5340472055382283e-06, + "loss": 1.0512, + "step": 3589 + }, + { + "epoch": 0.43167197739433655, + "grad_norm": 1.9537578320483537, + "learning_rate": 2.5332964778714468e-06, + "loss": 1.0505, + "step": 3590 + }, + { + "epoch": 0.43179222028497566, + "grad_norm": 1.8597659172918406, + "learning_rate": 2.5325456693040123e-06, + "loss": 0.9017, + "step": 3591 + }, + { + "epoch": 0.43191246317561477, + "grad_norm": 2.375365480108907, + "learning_rate": 2.531794779949824e-06, + "loss": 0.9845, + "step": 3592 + }, + { + "epoch": 0.4320327060662538, + "grad_norm": 1.842949709727213, + "learning_rate": 2.5310438099227903e-06, + "loss": 1.1181, + "step": 3593 + }, + { + "epoch": 0.43215294895689293, + "grad_norm": 0.8514222195773318, + "learning_rate": 2.530292759336833e-06, + "loss": 0.7716, + "step": 3594 + }, + { + "epoch": 0.432273191847532, + "grad_norm": 2.254027047713432, + "learning_rate": 2.5295416283058855e-06, + "loss": 0.9275, + "step": 3595 + }, + { + "epoch": 0.4323934347381711, + "grad_norm": 1.6015884025153833, + "learning_rate": 2.5287904169438943e-06, + "loss": 0.8969, + "step": 3596 + }, + { + "epoch": 0.4325136776288102, + "grad_norm": 2.5049322112270858, + "learning_rate": 2.528039125364817e-06, + "loss": 0.8812, + "step": 3597 + }, + { + "epoch": 0.43263392051944927, + "grad_norm": 2.139541908861088, + "learning_rate": 2.5272877536826246e-06, + "loss": 0.9996, + "step": 3598 + }, + { + "epoch": 0.4327541634100884, + "grad_norm": 2.2859554099726127, + "learning_rate": 2.5265363020112986e-06, + "loss": 0.946, + "step": 3599 + }, + { + "epoch": 0.4328744063007275, + "grad_norm": 1.8623286277220399, + "learning_rate": 2.5257847704648344e-06, + "loss": 1.0755, + "step": 3600 + }, + { + "epoch": 0.43299464919136654, + "grad_norm": 2.166562065631774, + "learning_rate": 2.525033159157239e-06, + "loss": 0.9909, + "step": 3601 + }, + { + "epoch": 0.43311489208200565, + "grad_norm": 1.935001485069947, + "learning_rate": 2.52428146820253e-06, + "loss": 1.0089, + "step": 3602 + }, + { + "epoch": 0.43323513497264476, + "grad_norm": 1.8164700404625311, + "learning_rate": 2.52352969771474e-06, + "loss": 1.0531, + "step": 3603 + }, + { + "epoch": 0.4333553778632838, + "grad_norm": 4.340079314038908, + "learning_rate": 2.5227778478079106e-06, + "loss": 1.1215, + "step": 3604 + }, + { + "epoch": 0.43347562075392293, + "grad_norm": 1.5678165014047487, + "learning_rate": 2.522025918596098e-06, + "loss": 1.0062, + "step": 3605 + }, + { + "epoch": 0.43359586364456204, + "grad_norm": 1.4551064712528374, + "learning_rate": 2.521273910193368e-06, + "loss": 0.8942, + "step": 3606 + }, + { + "epoch": 0.4337161065352011, + "grad_norm": 2.047152667831629, + "learning_rate": 2.5205218227138006e-06, + "loss": 1.1117, + "step": 3607 + }, + { + "epoch": 0.4338363494258402, + "grad_norm": 2.1033594326731553, + "learning_rate": 2.519769656271486e-06, + "loss": 1.0322, + "step": 3608 + }, + { + "epoch": 0.43395659231647926, + "grad_norm": 4.847585820524843, + "learning_rate": 2.5190174109805285e-06, + "loss": 0.9154, + "step": 3609 + }, + { + "epoch": 0.43407683520711837, + "grad_norm": 2.473400901769469, + "learning_rate": 2.518265086955042e-06, + "loss": 0.8849, + "step": 3610 + }, + { + "epoch": 0.4341970780977575, + "grad_norm": 1.8225580207788241, + "learning_rate": 2.5175126843091534e-06, + "loss": 1.0753, + "step": 3611 + }, + { + "epoch": 0.43431732098839654, + "grad_norm": 2.145741605804248, + "learning_rate": 2.5167602031570034e-06, + "loss": 0.9938, + "step": 3612 + }, + { + "epoch": 0.43443756387903565, + "grad_norm": 1.6596877211102106, + "learning_rate": 2.51600764361274e-06, + "loss": 0.971, + "step": 3613 + }, + { + "epoch": 0.43455780676967476, + "grad_norm": 2.780109728973491, + "learning_rate": 2.5152550057905283e-06, + "loss": 1.0247, + "step": 3614 + }, + { + "epoch": 0.4346780496603138, + "grad_norm": 2.046771796733293, + "learning_rate": 2.5145022898045415e-06, + "loss": 1.0055, + "step": 3615 + }, + { + "epoch": 0.4347982925509529, + "grad_norm": 2.0657861136020057, + "learning_rate": 2.5137494957689664e-06, + "loss": 1.1385, + "step": 3616 + }, + { + "epoch": 0.43491853544159204, + "grad_norm": 0.7754193740242648, + "learning_rate": 2.5129966237980016e-06, + "loss": 0.8117, + "step": 3617 + }, + { + "epoch": 0.4350387783322311, + "grad_norm": 2.2074350173388857, + "learning_rate": 2.512243674005857e-06, + "loss": 1.0182, + "step": 3618 + }, + { + "epoch": 0.4351590212228702, + "grad_norm": 1.7559351411455002, + "learning_rate": 2.5114906465067537e-06, + "loss": 1.0976, + "step": 3619 + }, + { + "epoch": 0.4352792641135093, + "grad_norm": 2.2395763544964544, + "learning_rate": 2.5107375414149264e-06, + "loss": 0.9858, + "step": 3620 + }, + { + "epoch": 0.43539950700414837, + "grad_norm": 2.50191148161943, + "learning_rate": 2.5099843588446197e-06, + "loss": 0.9574, + "step": 3621 + }, + { + "epoch": 0.4355197498947875, + "grad_norm": 1.6204098649232472, + "learning_rate": 2.509231098910091e-06, + "loss": 0.8525, + "step": 3622 + }, + { + "epoch": 0.4356399927854266, + "grad_norm": 2.183975322870426, + "learning_rate": 2.508477761725611e-06, + "loss": 0.982, + "step": 3623 + }, + { + "epoch": 0.43576023567606564, + "grad_norm": 1.7675435268324897, + "learning_rate": 2.507724347405458e-06, + "loss": 1.0487, + "step": 3624 + }, + { + "epoch": 0.43588047856670475, + "grad_norm": 2.3250531918583506, + "learning_rate": 2.5069708560639243e-06, + "loss": 1.063, + "step": 3625 + }, + { + "epoch": 0.4360007214573438, + "grad_norm": 2.0465444050325465, + "learning_rate": 2.5062172878153158e-06, + "loss": 0.8507, + "step": 3626 + }, + { + "epoch": 0.4361209643479829, + "grad_norm": 2.8391883230171837, + "learning_rate": 2.505463642773947e-06, + "loss": 1.1135, + "step": 3627 + }, + { + "epoch": 0.43624120723862203, + "grad_norm": 3.5294026631817617, + "learning_rate": 2.504709921054146e-06, + "loss": 0.9858, + "step": 3628 + }, + { + "epoch": 0.4363614501292611, + "grad_norm": 2.068385310456199, + "learning_rate": 2.50395612277025e-06, + "loss": 1.0698, + "step": 3629 + }, + { + "epoch": 0.4364816930199002, + "grad_norm": 2.085139775702413, + "learning_rate": 2.503202248036612e-06, + "loss": 0.9684, + "step": 3630 + }, + { + "epoch": 0.4366019359105393, + "grad_norm": 1.9527508186491966, + "learning_rate": 2.5024482969675927e-06, + "loss": 0.9724, + "step": 3631 + }, + { + "epoch": 0.43672217880117836, + "grad_norm": 2.2458053270998857, + "learning_rate": 2.501694269677566e-06, + "loss": 1.0807, + "step": 3632 + }, + { + "epoch": 0.4368424216918175, + "grad_norm": 3.8486017140252695, + "learning_rate": 2.500940166280918e-06, + "loss": 1.0428, + "step": 3633 + }, + { + "epoch": 0.4369626645824566, + "grad_norm": 2.0149660437967207, + "learning_rate": 2.500185986892045e-06, + "loss": 1.0224, + "step": 3634 + }, + { + "epoch": 0.43708290747309564, + "grad_norm": 2.1775578859257947, + "learning_rate": 2.499431731625355e-06, + "loss": 1.0111, + "step": 3635 + }, + { + "epoch": 0.43720315036373475, + "grad_norm": 2.001552098743771, + "learning_rate": 2.4986774005952686e-06, + "loss": 1.0319, + "step": 3636 + }, + { + "epoch": 0.43732339325437386, + "grad_norm": 2.2017285405526676, + "learning_rate": 2.4979229939162166e-06, + "loss": 1.0832, + "step": 3637 + }, + { + "epoch": 0.4374436361450129, + "grad_norm": 1.6426090414469332, + "learning_rate": 2.4971685117026433e-06, + "loss": 1.0432, + "step": 3638 + }, + { + "epoch": 0.437563879035652, + "grad_norm": 1.4609379079889555, + "learning_rate": 2.4964139540690018e-06, + "loss": 1.0039, + "step": 3639 + }, + { + "epoch": 0.4376841219262911, + "grad_norm": 2.994441923644877, + "learning_rate": 2.495659321129758e-06, + "loss": 0.9651, + "step": 3640 + }, + { + "epoch": 0.4378043648169302, + "grad_norm": 1.7902068763400545, + "learning_rate": 2.494904612999389e-06, + "loss": 0.9897, + "step": 3641 + }, + { + "epoch": 0.4379246077075693, + "grad_norm": 0.7807193861525396, + "learning_rate": 2.4941498297923843e-06, + "loss": 0.8033, + "step": 3642 + }, + { + "epoch": 0.43804485059820836, + "grad_norm": 1.7642687424416499, + "learning_rate": 2.4933949716232424e-06, + "loss": 0.9318, + "step": 3643 + }, + { + "epoch": 0.43816509348884747, + "grad_norm": 2.350590501819675, + "learning_rate": 2.492640038606476e-06, + "loss": 0.9741, + "step": 3644 + }, + { + "epoch": 0.4382853363794866, + "grad_norm": 1.9427987521677486, + "learning_rate": 2.491885030856608e-06, + "loss": 1.0195, + "step": 3645 + }, + { + "epoch": 0.43840557927012563, + "grad_norm": 2.16681757425861, + "learning_rate": 2.4911299484881713e-06, + "loss": 1.0641, + "step": 3646 + }, + { + "epoch": 0.43852582216076474, + "grad_norm": 1.662415670106076, + "learning_rate": 2.490374791615712e-06, + "loss": 1.0486, + "step": 3647 + }, + { + "epoch": 0.43864606505140386, + "grad_norm": 2.676778031293349, + "learning_rate": 2.4896195603537867e-06, + "loss": 1.0138, + "step": 3648 + }, + { + "epoch": 0.4387663079420429, + "grad_norm": 1.9374624679375723, + "learning_rate": 2.488864254816964e-06, + "loss": 0.9782, + "step": 3649 + }, + { + "epoch": 0.438886550832682, + "grad_norm": 3.2649333896823225, + "learning_rate": 2.4881088751198218e-06, + "loss": 0.9222, + "step": 3650 + }, + { + "epoch": 0.43900679372332113, + "grad_norm": 2.7453221502769525, + "learning_rate": 2.4873534213769517e-06, + "loss": 0.8814, + "step": 3651 + }, + { + "epoch": 0.4391270366139602, + "grad_norm": 1.9222024778161786, + "learning_rate": 2.4865978937029547e-06, + "loss": 0.9526, + "step": 3652 + }, + { + "epoch": 0.4392472795045993, + "grad_norm": 1.5842909510362673, + "learning_rate": 2.485842292212445e-06, + "loss": 0.9047, + "step": 3653 + }, + { + "epoch": 0.4393675223952384, + "grad_norm": 1.8461209586160796, + "learning_rate": 2.485086617020045e-06, + "loss": 1.0444, + "step": 3654 + }, + { + "epoch": 0.43948776528587746, + "grad_norm": 3.6856946243403836, + "learning_rate": 2.4843308682403903e-06, + "loss": 1.0525, + "step": 3655 + }, + { + "epoch": 0.4396080081765166, + "grad_norm": 5.691915621509406, + "learning_rate": 2.4835750459881294e-06, + "loss": 1.0663, + "step": 3656 + }, + { + "epoch": 0.43972825106715563, + "grad_norm": 1.9584534249785106, + "learning_rate": 2.4828191503779177e-06, + "loss": 1.0569, + "step": 3657 + }, + { + "epoch": 0.43984849395779474, + "grad_norm": 1.9725569596183017, + "learning_rate": 2.482063181524425e-06, + "loss": 1.1354, + "step": 3658 + }, + { + "epoch": 0.43996873684843385, + "grad_norm": 2.4535042050537244, + "learning_rate": 2.4813071395423307e-06, + "loss": 1.0472, + "step": 3659 + }, + { + "epoch": 0.4400889797390729, + "grad_norm": 2.565528382725157, + "learning_rate": 2.4805510245463263e-06, + "loss": 0.8832, + "step": 3660 + }, + { + "epoch": 0.440209222629712, + "grad_norm": 2.0838489148805457, + "learning_rate": 2.4797948366511137e-06, + "loss": 0.8404, + "step": 3661 + }, + { + "epoch": 0.4403294655203511, + "grad_norm": 2.057655656379836, + "learning_rate": 2.4790385759714055e-06, + "loss": 0.9968, + "step": 3662 + }, + { + "epoch": 0.4404497084109902, + "grad_norm": 1.7743253795945488, + "learning_rate": 2.478282242621926e-06, + "loss": 0.9482, + "step": 3663 + }, + { + "epoch": 0.4405699513016293, + "grad_norm": 0.8404540926975536, + "learning_rate": 2.477525836717411e-06, + "loss": 0.8326, + "step": 3664 + }, + { + "epoch": 0.4406901941922684, + "grad_norm": 2.099998714809933, + "learning_rate": 2.476769358372606e-06, + "loss": 1.034, + "step": 3665 + }, + { + "epoch": 0.44081043708290746, + "grad_norm": 2.2644858818060505, + "learning_rate": 2.4760128077022683e-06, + "loss": 0.9886, + "step": 3666 + }, + { + "epoch": 0.44093067997354657, + "grad_norm": 1.5168560573556649, + "learning_rate": 2.4752561848211672e-06, + "loss": 0.9215, + "step": 3667 + }, + { + "epoch": 0.4410509228641857, + "grad_norm": 3.007542823492966, + "learning_rate": 2.4744994898440797e-06, + "loss": 0.9526, + "step": 3668 + }, + { + "epoch": 0.44117116575482473, + "grad_norm": 2.0288018365373333, + "learning_rate": 2.473742722885797e-06, + "loss": 1.0718, + "step": 3669 + }, + { + "epoch": 0.44129140864546385, + "grad_norm": 2.016704536376669, + "learning_rate": 2.4729858840611197e-06, + "loss": 0.8866, + "step": 3670 + }, + { + "epoch": 0.4414116515361029, + "grad_norm": 2.2647969804601504, + "learning_rate": 2.4722289734848605e-06, + "loss": 0.9629, + "step": 3671 + }, + { + "epoch": 0.441531894426742, + "grad_norm": 1.9954696246888928, + "learning_rate": 2.471471991271841e-06, + "loss": 1.0179, + "step": 3672 + }, + { + "epoch": 0.4416521373173811, + "grad_norm": 2.058135761657993, + "learning_rate": 2.470714937536896e-06, + "loss": 1.0332, + "step": 3673 + }, + { + "epoch": 0.4417723802080202, + "grad_norm": 2.002582670641261, + "learning_rate": 2.469957812394868e-06, + "loss": 0.9426, + "step": 3674 + }, + { + "epoch": 0.4418926230986593, + "grad_norm": 2.018007038977514, + "learning_rate": 2.4692006159606148e-06, + "loss": 1.0007, + "step": 3675 + }, + { + "epoch": 0.4420128659892984, + "grad_norm": 1.923499436547217, + "learning_rate": 2.468443348349e-06, + "loss": 1.0243, + "step": 3676 + }, + { + "epoch": 0.44213310887993745, + "grad_norm": 3.7800041226965004, + "learning_rate": 2.467686009674902e-06, + "loss": 1.0649, + "step": 3677 + }, + { + "epoch": 0.44225335177057656, + "grad_norm": 2.03859228800001, + "learning_rate": 2.466928600053209e-06, + "loss": 1.0891, + "step": 3678 + }, + { + "epoch": 0.4423735946612157, + "grad_norm": 1.7740662254306476, + "learning_rate": 2.466171119598818e-06, + "loss": 0.9503, + "step": 3679 + }, + { + "epoch": 0.44249383755185473, + "grad_norm": 2.1156140776539627, + "learning_rate": 2.465413568426639e-06, + "loss": 1.0085, + "step": 3680 + }, + { + "epoch": 0.44261408044249384, + "grad_norm": 2.5908890537843905, + "learning_rate": 2.464655946651591e-06, + "loss": 1.0497, + "step": 3681 + }, + { + "epoch": 0.44273432333313295, + "grad_norm": 1.965464802871551, + "learning_rate": 2.4638982543886065e-06, + "loss": 1.046, + "step": 3682 + }, + { + "epoch": 0.442854566223772, + "grad_norm": 2.4804799958221944, + "learning_rate": 2.4631404917526254e-06, + "loss": 1.1109, + "step": 3683 + }, + { + "epoch": 0.4429748091144111, + "grad_norm": 1.8210045804547976, + "learning_rate": 2.4623826588586e-06, + "loss": 1.0306, + "step": 3684 + }, + { + "epoch": 0.4430950520050502, + "grad_norm": 1.6089800007337662, + "learning_rate": 2.461624755821492e-06, + "loss": 1.0634, + "step": 3685 + }, + { + "epoch": 0.4432152948956893, + "grad_norm": 1.8527232205760706, + "learning_rate": 2.4608667827562763e-06, + "loss": 1.0003, + "step": 3686 + }, + { + "epoch": 0.4433355377863284, + "grad_norm": 2.0518584410562513, + "learning_rate": 2.460108739777936e-06, + "loss": 1.1403, + "step": 3687 + }, + { + "epoch": 0.44345578067696745, + "grad_norm": 1.7930682119403318, + "learning_rate": 2.4593506270014656e-06, + "loss": 1.0026, + "step": 3688 + }, + { + "epoch": 0.44357602356760656, + "grad_norm": 1.7872024928779942, + "learning_rate": 2.45859244454187e-06, + "loss": 1.0579, + "step": 3689 + }, + { + "epoch": 0.44369626645824567, + "grad_norm": 2.0802858060935234, + "learning_rate": 2.4578341925141655e-06, + "loss": 0.9009, + "step": 3690 + }, + { + "epoch": 0.4438165093488847, + "grad_norm": 2.2002752825430107, + "learning_rate": 2.457075871033378e-06, + "loss": 0.9593, + "step": 3691 + }, + { + "epoch": 0.44393675223952384, + "grad_norm": 2.393512453090258, + "learning_rate": 2.4563174802145445e-06, + "loss": 1.1229, + "step": 3692 + }, + { + "epoch": 0.44405699513016295, + "grad_norm": 0.6497298541261152, + "learning_rate": 2.455559020172712e-06, + "loss": 0.7238, + "step": 3693 + }, + { + "epoch": 0.444177238020802, + "grad_norm": 1.8835295898446387, + "learning_rate": 2.4548004910229385e-06, + "loss": 1.1382, + "step": 3694 + }, + { + "epoch": 0.4442974809114411, + "grad_norm": 2.063522085616519, + "learning_rate": 2.4540418928802913e-06, + "loss": 1.1082, + "step": 3695 + }, + { + "epoch": 0.4444177238020802, + "grad_norm": 2.3954024107022183, + "learning_rate": 2.4532832258598506e-06, + "loss": 0.8946, + "step": 3696 + }, + { + "epoch": 0.4445379666927193, + "grad_norm": 1.779843360566619, + "learning_rate": 2.4525244900767047e-06, + "loss": 1.0416, + "step": 3697 + }, + { + "epoch": 0.4446582095833584, + "grad_norm": 0.7727769284944905, + "learning_rate": 2.4517656856459536e-06, + "loss": 0.8427, + "step": 3698 + }, + { + "epoch": 0.4447784524739975, + "grad_norm": 1.9807311717473033, + "learning_rate": 2.4510068126827073e-06, + "loss": 0.9188, + "step": 3699 + }, + { + "epoch": 0.44489869536463655, + "grad_norm": 2.457449436230259, + "learning_rate": 2.450247871302086e-06, + "loss": 1.0553, + "step": 3700 + }, + { + "epoch": 0.44501893825527566, + "grad_norm": 2.4717338975065677, + "learning_rate": 2.44948886161922e-06, + "loss": 1.0705, + "step": 3701 + }, + { + "epoch": 0.4451391811459148, + "grad_norm": 1.7036145407884424, + "learning_rate": 2.4487297837492524e-06, + "loss": 1.0853, + "step": 3702 + }, + { + "epoch": 0.44525942403655383, + "grad_norm": 2.154342249706306, + "learning_rate": 2.4479706378073323e-06, + "loss": 0.8628, + "step": 3703 + }, + { + "epoch": 0.44537966692719294, + "grad_norm": 2.3032436179848284, + "learning_rate": 2.447211423908623e-06, + "loss": 1.0761, + "step": 3704 + }, + { + "epoch": 0.445499909817832, + "grad_norm": 2.1367638285064348, + "learning_rate": 2.4464521421682966e-06, + "loss": 0.9929, + "step": 3705 + }, + { + "epoch": 0.4456201527084711, + "grad_norm": 1.4156554296730346, + "learning_rate": 2.4456927927015345e-06, + "loss": 1.115, + "step": 3706 + }, + { + "epoch": 0.4457403955991102, + "grad_norm": 2.149039555202896, + "learning_rate": 2.4449333756235307e-06, + "loss": 1.0016, + "step": 3707 + }, + { + "epoch": 0.4458606384897493, + "grad_norm": 2.1649099955890696, + "learning_rate": 2.4441738910494876e-06, + "loss": 1.0263, + "step": 3708 + }, + { + "epoch": 0.4459808813803884, + "grad_norm": 2.072252969226788, + "learning_rate": 2.4434143390946176e-06, + "loss": 1.0605, + "step": 3709 + }, + { + "epoch": 0.4461011242710275, + "grad_norm": 1.95757914961673, + "learning_rate": 2.4426547198741457e-06, + "loss": 1.0923, + "step": 3710 + }, + { + "epoch": 0.44622136716166655, + "grad_norm": 2.3199463629277233, + "learning_rate": 2.441895033503305e-06, + "loss": 0.9806, + "step": 3711 + }, + { + "epoch": 0.44634161005230566, + "grad_norm": 1.8335233358703544, + "learning_rate": 2.4411352800973375e-06, + "loss": 1.0587, + "step": 3712 + }, + { + "epoch": 0.44646185294294477, + "grad_norm": 2.318605045198802, + "learning_rate": 2.4403754597715005e-06, + "loss": 0.9914, + "step": 3713 + }, + { + "epoch": 0.4465820958335838, + "grad_norm": 3.051313405141944, + "learning_rate": 2.4396155726410553e-06, + "loss": 1.1681, + "step": 3714 + }, + { + "epoch": 0.44670233872422294, + "grad_norm": 2.642891988561376, + "learning_rate": 2.438855618821278e-06, + "loss": 1.1452, + "step": 3715 + }, + { + "epoch": 0.44682258161486205, + "grad_norm": 1.7296296231477604, + "learning_rate": 2.4380955984274517e-06, + "loss": 0.9143, + "step": 3716 + }, + { + "epoch": 0.4469428245055011, + "grad_norm": 1.8112936280276575, + "learning_rate": 2.4373355115748716e-06, + "loss": 1.0105, + "step": 3717 + }, + { + "epoch": 0.4470630673961402, + "grad_norm": 1.7173812965118267, + "learning_rate": 2.436575358378842e-06, + "loss": 0.9588, + "step": 3718 + }, + { + "epoch": 0.44718331028677927, + "grad_norm": 2.573664743086859, + "learning_rate": 2.4358151389546782e-06, + "loss": 1.0734, + "step": 3719 + }, + { + "epoch": 0.4473035531774184, + "grad_norm": 2.586484177074044, + "learning_rate": 2.4350548534177035e-06, + "loss": 0.9999, + "step": 3720 + }, + { + "epoch": 0.4474237960680575, + "grad_norm": 20.036060630421776, + "learning_rate": 2.434294501883254e-06, + "loss": 0.9059, + "step": 3721 + }, + { + "epoch": 0.44754403895869654, + "grad_norm": 1.6811916270279663, + "learning_rate": 2.433534084466674e-06, + "loss": 0.8952, + "step": 3722 + }, + { + "epoch": 0.44766428184933565, + "grad_norm": 1.7848143135321004, + "learning_rate": 2.4327736012833178e-06, + "loss": 0.9445, + "step": 3723 + }, + { + "epoch": 0.44778452473997477, + "grad_norm": 2.9358889443288922, + "learning_rate": 2.4320130524485506e-06, + "loss": 1.0018, + "step": 3724 + }, + { + "epoch": 0.4479047676306138, + "grad_norm": 1.5762933355752833, + "learning_rate": 2.431252438077746e-06, + "loss": 1.0362, + "step": 3725 + }, + { + "epoch": 0.44802501052125293, + "grad_norm": 2.2162919734071953, + "learning_rate": 2.4304917582862906e-06, + "loss": 1.0107, + "step": 3726 + }, + { + "epoch": 0.44814525341189204, + "grad_norm": 2.1834311655077423, + "learning_rate": 2.4297310131895774e-06, + "loss": 1.1186, + "step": 3727 + }, + { + "epoch": 0.4482654963025311, + "grad_norm": 2.420538945991677, + "learning_rate": 2.4289702029030113e-06, + "loss": 0.9884, + "step": 3728 + }, + { + "epoch": 0.4483857391931702, + "grad_norm": 1.8047330776255612, + "learning_rate": 2.4282093275420057e-06, + "loss": 1.0644, + "step": 3729 + }, + { + "epoch": 0.4485059820838093, + "grad_norm": 2.4095402800673598, + "learning_rate": 2.4274483872219863e-06, + "loss": 0.9422, + "step": 3730 + }, + { + "epoch": 0.4486262249744484, + "grad_norm": 1.8088975161048073, + "learning_rate": 2.426687382058386e-06, + "loss": 1.1736, + "step": 3731 + }, + { + "epoch": 0.4487464678650875, + "grad_norm": 0.8745395606980055, + "learning_rate": 2.425926312166649e-06, + "loss": 0.8364, + "step": 3732 + }, + { + "epoch": 0.4488667107557266, + "grad_norm": 2.4836417981612176, + "learning_rate": 2.42516517766223e-06, + "loss": 0.9687, + "step": 3733 + }, + { + "epoch": 0.44898695364636565, + "grad_norm": 2.050971205860303, + "learning_rate": 2.4244039786605907e-06, + "loss": 0.9167, + "step": 3734 + }, + { + "epoch": 0.44910719653700476, + "grad_norm": 2.4205766704845546, + "learning_rate": 2.4236427152772055e-06, + "loss": 1.0627, + "step": 3735 + }, + { + "epoch": 0.4492274394276438, + "grad_norm": 0.8348699865943147, + "learning_rate": 2.422881387627557e-06, + "loss": 0.812, + "step": 3736 + }, + { + "epoch": 0.4493476823182829, + "grad_norm": 1.7144432903710864, + "learning_rate": 2.422119995827139e-06, + "loss": 1.0124, + "step": 3737 + }, + { + "epoch": 0.44946792520892204, + "grad_norm": 2.603782004875284, + "learning_rate": 2.4213585399914528e-06, + "loss": 0.9719, + "step": 3738 + }, + { + "epoch": 0.4495881680995611, + "grad_norm": 2.9411457245285546, + "learning_rate": 2.4205970202360113e-06, + "loss": 1.0891, + "step": 3739 + }, + { + "epoch": 0.4497084109902002, + "grad_norm": 2.686976803263861, + "learning_rate": 2.4198354366763354e-06, + "loss": 1.0194, + "step": 3740 + }, + { + "epoch": 0.4498286538808393, + "grad_norm": 1.9881053314627166, + "learning_rate": 2.4190737894279587e-06, + "loss": 1.0182, + "step": 3741 + }, + { + "epoch": 0.44994889677147837, + "grad_norm": 2.152009969262781, + "learning_rate": 2.4183120786064203e-06, + "loss": 1.0409, + "step": 3742 + }, + { + "epoch": 0.4500691396621175, + "grad_norm": 2.8687583507673935, + "learning_rate": 2.417550304327273e-06, + "loss": 1.0949, + "step": 3743 + }, + { + "epoch": 0.4501893825527566, + "grad_norm": 1.7154853333146713, + "learning_rate": 2.4167884667060763e-06, + "loss": 0.995, + "step": 3744 + }, + { + "epoch": 0.45030962544339564, + "grad_norm": 2.207532005553259, + "learning_rate": 2.4160265658584e-06, + "loss": 1.1106, + "step": 3745 + }, + { + "epoch": 0.45042986833403476, + "grad_norm": 2.1041012996176622, + "learning_rate": 2.4152646018998253e-06, + "loss": 0.9228, + "step": 3746 + }, + { + "epoch": 0.45055011122467387, + "grad_norm": 1.8076440942081722, + "learning_rate": 2.4145025749459403e-06, + "loss": 0.9534, + "step": 3747 + }, + { + "epoch": 0.4506703541153129, + "grad_norm": 1.9145492246271727, + "learning_rate": 2.413740485112344e-06, + "loss": 0.9313, + "step": 3748 + }, + { + "epoch": 0.45079059700595203, + "grad_norm": 1.6749373580834055, + "learning_rate": 2.412978332514646e-06, + "loss": 1.0607, + "step": 3749 + }, + { + "epoch": 0.4509108398965911, + "grad_norm": 2.4173504311528, + "learning_rate": 2.4122161172684623e-06, + "loss": 0.9573, + "step": 3750 + }, + { + "epoch": 0.4510310827872302, + "grad_norm": 2.1482654849461773, + "learning_rate": 2.4114538394894216e-06, + "loss": 1.0774, + "step": 3751 + }, + { + "epoch": 0.4511513256778693, + "grad_norm": 1.9414800556974936, + "learning_rate": 2.410691499293161e-06, + "loss": 1.0646, + "step": 3752 + }, + { + "epoch": 0.45127156856850836, + "grad_norm": 1.5852592032489816, + "learning_rate": 2.409929096795326e-06, + "loss": 0.975, + "step": 3753 + }, + { + "epoch": 0.4513918114591475, + "grad_norm": 2.0792031610480275, + "learning_rate": 2.409166632111573e-06, + "loss": 1.0292, + "step": 3754 + }, + { + "epoch": 0.4515120543497866, + "grad_norm": 2.0196110545462727, + "learning_rate": 2.4084041053575674e-06, + "loss": 1.0425, + "step": 3755 + }, + { + "epoch": 0.45163229724042564, + "grad_norm": 1.9398799553333514, + "learning_rate": 2.4076415166489834e-06, + "loss": 0.9635, + "step": 3756 + }, + { + "epoch": 0.45175254013106475, + "grad_norm": 1.6196950738528482, + "learning_rate": 2.406878866101506e-06, + "loss": 1.0284, + "step": 3757 + }, + { + "epoch": 0.45187278302170386, + "grad_norm": 2.1288249428997696, + "learning_rate": 2.4061161538308273e-06, + "loss": 1.0179, + "step": 3758 + }, + { + "epoch": 0.4519930259123429, + "grad_norm": 1.895325798874857, + "learning_rate": 2.4053533799526523e-06, + "loss": 1.1298, + "step": 3759 + }, + { + "epoch": 0.452113268802982, + "grad_norm": 1.9179486120219524, + "learning_rate": 2.404590544582691e-06, + "loss": 1.1009, + "step": 3760 + }, + { + "epoch": 0.45223351169362114, + "grad_norm": 1.6409772222357275, + "learning_rate": 2.403827647836666e-06, + "loss": 1.048, + "step": 3761 + }, + { + "epoch": 0.4523537545842602, + "grad_norm": 2.5492292885032035, + "learning_rate": 2.4030646898303075e-06, + "loss": 0.9327, + "step": 3762 + }, + { + "epoch": 0.4524739974748993, + "grad_norm": 2.3665007045026676, + "learning_rate": 2.4023016706793566e-06, + "loss": 1.0551, + "step": 3763 + }, + { + "epoch": 0.4525942403655384, + "grad_norm": 0.7759587203038184, + "learning_rate": 2.401538590499561e-06, + "loss": 0.8065, + "step": 3764 + }, + { + "epoch": 0.45271448325617747, + "grad_norm": 2.1383804471104932, + "learning_rate": 2.400775449406682e-06, + "loss": 0.9544, + "step": 3765 + }, + { + "epoch": 0.4528347261468166, + "grad_norm": 1.853811515770033, + "learning_rate": 2.400012247516485e-06, + "loss": 0.9689, + "step": 3766 + }, + { + "epoch": 0.45295496903745563, + "grad_norm": 1.729505352571117, + "learning_rate": 2.3992489849447484e-06, + "loss": 1.1435, + "step": 3767 + }, + { + "epoch": 0.45307521192809475, + "grad_norm": 1.8593862196639896, + "learning_rate": 2.3984856618072584e-06, + "loss": 1.0271, + "step": 3768 + }, + { + "epoch": 0.45319545481873386, + "grad_norm": 1.8915042566609013, + "learning_rate": 2.3977222782198098e-06, + "loss": 0.9804, + "step": 3769 + }, + { + "epoch": 0.4533156977093729, + "grad_norm": 2.129523455542361, + "learning_rate": 2.3969588342982077e-06, + "loss": 0.9917, + "step": 3770 + }, + { + "epoch": 0.453435940600012, + "grad_norm": 1.6138089035240528, + "learning_rate": 2.396195330158267e-06, + "loss": 0.9616, + "step": 3771 + }, + { + "epoch": 0.45355618349065113, + "grad_norm": 2.008362097422069, + "learning_rate": 2.3954317659158094e-06, + "loss": 1.0342, + "step": 3772 + }, + { + "epoch": 0.4536764263812902, + "grad_norm": 0.8928996702692819, + "learning_rate": 2.394668141686667e-06, + "loss": 0.8066, + "step": 3773 + }, + { + "epoch": 0.4537966692719293, + "grad_norm": 1.9915425172788233, + "learning_rate": 2.3939044575866813e-06, + "loss": 0.938, + "step": 3774 + }, + { + "epoch": 0.4539169121625684, + "grad_norm": 2.299800748072528, + "learning_rate": 2.3931407137317024e-06, + "loss": 0.9913, + "step": 3775 + }, + { + "epoch": 0.45403715505320746, + "grad_norm": 1.84020338168004, + "learning_rate": 2.3923769102375907e-06, + "loss": 1.0842, + "step": 3776 + }, + { + "epoch": 0.4541573979438466, + "grad_norm": 2.614138703452568, + "learning_rate": 2.391613047220213e-06, + "loss": 1.0216, + "step": 3777 + }, + { + "epoch": 0.4542776408344857, + "grad_norm": 2.434090699818441, + "learning_rate": 2.390849124795447e-06, + "loss": 1.0311, + "step": 3778 + }, + { + "epoch": 0.45439788372512474, + "grad_norm": 2.079498066862051, + "learning_rate": 2.3900851430791804e-06, + "loss": 1.0795, + "step": 3779 + }, + { + "epoch": 0.45451812661576385, + "grad_norm": 2.187288764563884, + "learning_rate": 2.389321102187307e-06, + "loss": 1.0852, + "step": 3780 + }, + { + "epoch": 0.4546383695064029, + "grad_norm": 1.9619697194538226, + "learning_rate": 2.3885570022357326e-06, + "loss": 1.0553, + "step": 3781 + }, + { + "epoch": 0.454758612397042, + "grad_norm": 0.8140046053341393, + "learning_rate": 2.38779284334037e-06, + "loss": 0.8467, + "step": 3782 + }, + { + "epoch": 0.4548788552876811, + "grad_norm": 2.0827888413020634, + "learning_rate": 2.387028625617141e-06, + "loss": 1.0292, + "step": 3783 + }, + { + "epoch": 0.4549990981783202, + "grad_norm": 2.083817565591451, + "learning_rate": 2.3862643491819766e-06, + "loss": 1.0855, + "step": 3784 + }, + { + "epoch": 0.4551193410689593, + "grad_norm": 1.707498935905784, + "learning_rate": 2.3855000141508186e-06, + "loss": 1.0802, + "step": 3785 + }, + { + "epoch": 0.4552395839595984, + "grad_norm": 2.334627530534442, + "learning_rate": 2.3847356206396143e-06, + "loss": 1.0794, + "step": 3786 + }, + { + "epoch": 0.45535982685023746, + "grad_norm": 1.781080271347532, + "learning_rate": 2.3839711687643227e-06, + "loss": 1.0239, + "step": 3787 + }, + { + "epoch": 0.45548006974087657, + "grad_norm": 1.9180484296792328, + "learning_rate": 2.38320665864091e-06, + "loss": 0.9766, + "step": 3788 + }, + { + "epoch": 0.4556003126315157, + "grad_norm": 1.7630776452997505, + "learning_rate": 2.3824420903853516e-06, + "loss": 1.0584, + "step": 3789 + }, + { + "epoch": 0.45572055552215474, + "grad_norm": 2.277455767171635, + "learning_rate": 2.3816774641136324e-06, + "loss": 1.0592, + "step": 3790 + }, + { + "epoch": 0.45584079841279385, + "grad_norm": 2.042956612953909, + "learning_rate": 2.380912779941745e-06, + "loss": 0.9501, + "step": 3791 + }, + { + "epoch": 0.45596104130343296, + "grad_norm": 2.107603332230777, + "learning_rate": 2.3801480379856918e-06, + "loss": 1.0675, + "step": 3792 + }, + { + "epoch": 0.456081284194072, + "grad_norm": 1.7074701718228622, + "learning_rate": 2.379383238361484e-06, + "loss": 1.0701, + "step": 3793 + }, + { + "epoch": 0.4562015270847111, + "grad_norm": 2.0063971492245147, + "learning_rate": 2.3786183811851407e-06, + "loss": 1.0378, + "step": 3794 + }, + { + "epoch": 0.45632176997535023, + "grad_norm": 1.7579347016984992, + "learning_rate": 2.3778534665726892e-06, + "loss": 1.0405, + "step": 3795 + }, + { + "epoch": 0.4564420128659893, + "grad_norm": 2.106168329936404, + "learning_rate": 2.377088494640168e-06, + "loss": 0.9612, + "step": 3796 + }, + { + "epoch": 0.4565622557566284, + "grad_norm": 1.7966372125271979, + "learning_rate": 2.3763234655036216e-06, + "loss": 1.0191, + "step": 3797 + }, + { + "epoch": 0.45668249864726745, + "grad_norm": 7.580202175856226, + "learning_rate": 2.3755583792791046e-06, + "loss": 1.1103, + "step": 3798 + }, + { + "epoch": 0.45680274153790656, + "grad_norm": 2.343436563514271, + "learning_rate": 2.3747932360826803e-06, + "loss": 0.9791, + "step": 3799 + }, + { + "epoch": 0.4569229844285457, + "grad_norm": 2.0080954504465787, + "learning_rate": 2.3740280360304205e-06, + "loss": 1.0596, + "step": 3800 + }, + { + "epoch": 0.45704322731918473, + "grad_norm": 1.9385217003167328, + "learning_rate": 2.3732627792384038e-06, + "loss": 0.9168, + "step": 3801 + }, + { + "epoch": 0.45716347020982384, + "grad_norm": 2.1527695139517826, + "learning_rate": 2.3724974658227207e-06, + "loss": 0.989, + "step": 3802 + }, + { + "epoch": 0.45728371310046295, + "grad_norm": 1.9193356493111873, + "learning_rate": 2.3717320958994687e-06, + "loss": 0.9467, + "step": 3803 + }, + { + "epoch": 0.457403955991102, + "grad_norm": 2.210761181166475, + "learning_rate": 2.3709666695847534e-06, + "loss": 0.942, + "step": 3804 + }, + { + "epoch": 0.4575241988817411, + "grad_norm": 1.7863907500033098, + "learning_rate": 2.370201186994689e-06, + "loss": 0.9415, + "step": 3805 + }, + { + "epoch": 0.45764444177238023, + "grad_norm": 1.926111900929207, + "learning_rate": 2.369435648245399e-06, + "loss": 0.9347, + "step": 3806 + }, + { + "epoch": 0.4577646846630193, + "grad_norm": 1.9813202902662783, + "learning_rate": 2.368670053453015e-06, + "loss": 1.0848, + "step": 3807 + }, + { + "epoch": 0.4578849275536584, + "grad_norm": 2.162573126289674, + "learning_rate": 2.3679044027336757e-06, + "loss": 0.9804, + "step": 3808 + }, + { + "epoch": 0.4580051704442975, + "grad_norm": 3.0587633653480473, + "learning_rate": 2.3671386962035326e-06, + "loss": 0.9273, + "step": 3809 + }, + { + "epoch": 0.45812541333493656, + "grad_norm": 2.0802372114927428, + "learning_rate": 2.3663729339787405e-06, + "loss": 0.9195, + "step": 3810 + }, + { + "epoch": 0.45824565622557567, + "grad_norm": 2.5079496827439614, + "learning_rate": 2.365607116175466e-06, + "loss": 0.9749, + "step": 3811 + }, + { + "epoch": 0.4583658991162148, + "grad_norm": 2.384154795306332, + "learning_rate": 2.3648412429098825e-06, + "loss": 0.9067, + "step": 3812 + }, + { + "epoch": 0.45848614200685384, + "grad_norm": 1.9848811187505688, + "learning_rate": 2.364075314298172e-06, + "loss": 1.06, + "step": 3813 + }, + { + "epoch": 0.45860638489749295, + "grad_norm": 4.66762489744276, + "learning_rate": 2.3633093304565267e-06, + "loss": 0.9394, + "step": 3814 + }, + { + "epoch": 0.458726627788132, + "grad_norm": 1.8180631566340069, + "learning_rate": 2.3625432915011443e-06, + "loss": 0.8681, + "step": 3815 + }, + { + "epoch": 0.4588468706787711, + "grad_norm": 1.8491977447103238, + "learning_rate": 2.3617771975482334e-06, + "loss": 0.8905, + "step": 3816 + }, + { + "epoch": 0.4589671135694102, + "grad_norm": 1.9623844234182706, + "learning_rate": 2.3610110487140083e-06, + "loss": 0.9812, + "step": 3817 + }, + { + "epoch": 0.4590873564600493, + "grad_norm": 1.792831914330167, + "learning_rate": 2.360244845114695e-06, + "loss": 1.043, + "step": 3818 + }, + { + "epoch": 0.4592075993506884, + "grad_norm": 2.3183619459824096, + "learning_rate": 2.3594785868665245e-06, + "loss": 0.9253, + "step": 3819 + }, + { + "epoch": 0.4593278422413275, + "grad_norm": 2.33322935780969, + "learning_rate": 2.3587122740857386e-06, + "loss": 1.0412, + "step": 3820 + }, + { + "epoch": 0.45944808513196655, + "grad_norm": 2.1486962734782176, + "learning_rate": 2.357945906888586e-06, + "loss": 1.0188, + "step": 3821 + }, + { + "epoch": 0.45956832802260567, + "grad_norm": 2.291655754294568, + "learning_rate": 2.357179485391324e-06, + "loss": 1.0344, + "step": 3822 + }, + { + "epoch": 0.4596885709132448, + "grad_norm": 1.9070486678187957, + "learning_rate": 2.3564130097102173e-06, + "loss": 1.0982, + "step": 3823 + }, + { + "epoch": 0.45980881380388383, + "grad_norm": 1.7028195868512848, + "learning_rate": 2.355646479961541e-06, + "loss": 0.9891, + "step": 3824 + }, + { + "epoch": 0.45992905669452294, + "grad_norm": 2.006958538682045, + "learning_rate": 2.354879896261576e-06, + "loss": 0.9548, + "step": 3825 + }, + { + "epoch": 0.46004929958516205, + "grad_norm": 1.9424641296334715, + "learning_rate": 2.3541132587266133e-06, + "loss": 0.8071, + "step": 3826 + }, + { + "epoch": 0.4601695424758011, + "grad_norm": 1.922426594298864, + "learning_rate": 2.3533465674729515e-06, + "loss": 0.9295, + "step": 3827 + }, + { + "epoch": 0.4602897853664402, + "grad_norm": 2.068270269821403, + "learning_rate": 2.352579822616895e-06, + "loss": 0.9688, + "step": 3828 + }, + { + "epoch": 0.4604100282570793, + "grad_norm": 1.8039521539860335, + "learning_rate": 2.351813024274761e-06, + "loss": 1.0147, + "step": 3829 + }, + { + "epoch": 0.4605302711477184, + "grad_norm": 1.7640769721384468, + "learning_rate": 2.3510461725628693e-06, + "loss": 0.9722, + "step": 3830 + }, + { + "epoch": 0.4606505140383575, + "grad_norm": 1.7844438529551343, + "learning_rate": 2.350279267597554e-06, + "loss": 0.946, + "step": 3831 + }, + { + "epoch": 0.46077075692899655, + "grad_norm": 2.390004794383328, + "learning_rate": 2.3495123094951515e-06, + "loss": 1.0657, + "step": 3832 + }, + { + "epoch": 0.46089099981963566, + "grad_norm": 2.1137879728383853, + "learning_rate": 2.34874529837201e-06, + "loss": 0.9958, + "step": 3833 + }, + { + "epoch": 0.46101124271027477, + "grad_norm": 5.252404343588735, + "learning_rate": 2.347978234344483e-06, + "loss": 1.0233, + "step": 3834 + }, + { + "epoch": 0.4611314856009138, + "grad_norm": 1.7145478638525686, + "learning_rate": 2.347211117528935e-06, + "loss": 0.9311, + "step": 3835 + }, + { + "epoch": 0.46125172849155294, + "grad_norm": 1.6866047921849867, + "learning_rate": 2.3464439480417374e-06, + "loss": 0.9569, + "step": 3836 + }, + { + "epoch": 0.46137197138219205, + "grad_norm": 3.7537494987921898, + "learning_rate": 2.3456767259992676e-06, + "loss": 1.007, + "step": 3837 + }, + { + "epoch": 0.4614922142728311, + "grad_norm": 2.2558711064021257, + "learning_rate": 2.3449094515179135e-06, + "loss": 1.1305, + "step": 3838 + }, + { + "epoch": 0.4616124571634702, + "grad_norm": 1.6330427527639122, + "learning_rate": 2.34414212471407e-06, + "loss": 1.0555, + "step": 3839 + }, + { + "epoch": 0.4617327000541093, + "grad_norm": 1.9786128691474263, + "learning_rate": 2.3433747457041394e-06, + "loss": 0.9644, + "step": 3840 + }, + { + "epoch": 0.4618529429447484, + "grad_norm": 2.038350415765148, + "learning_rate": 2.342607314604533e-06, + "loss": 1.0799, + "step": 3841 + }, + { + "epoch": 0.4619731858353875, + "grad_norm": 1.7339952070220563, + "learning_rate": 2.3418398315316694e-06, + "loss": 1.0809, + "step": 3842 + }, + { + "epoch": 0.4620934287260266, + "grad_norm": 3.4669019570341906, + "learning_rate": 2.3410722966019755e-06, + "loss": 1.0218, + "step": 3843 + }, + { + "epoch": 0.46221367161666566, + "grad_norm": 1.793003322957795, + "learning_rate": 2.3403047099318848e-06, + "loss": 0.8918, + "step": 3844 + }, + { + "epoch": 0.46233391450730477, + "grad_norm": 2.3404901850216895, + "learning_rate": 2.3395370716378405e-06, + "loss": 0.9893, + "step": 3845 + }, + { + "epoch": 0.4624541573979438, + "grad_norm": 2.1629660131823916, + "learning_rate": 2.338769381836292e-06, + "loss": 0.9594, + "step": 3846 + }, + { + "epoch": 0.46257440028858293, + "grad_norm": 1.9965097848569797, + "learning_rate": 2.3380016406436984e-06, + "loss": 0.9652, + "step": 3847 + }, + { + "epoch": 0.46269464317922204, + "grad_norm": 1.953939405401736, + "learning_rate": 2.337233848176524e-06, + "loss": 1.0501, + "step": 3848 + }, + { + "epoch": 0.4628148860698611, + "grad_norm": 1.9816740382479574, + "learning_rate": 2.3364660045512435e-06, + "loss": 1.0712, + "step": 3849 + }, + { + "epoch": 0.4629351289605002, + "grad_norm": 0.7442721353301043, + "learning_rate": 2.335698109884337e-06, + "loss": 0.8196, + "step": 3850 + }, + { + "epoch": 0.4630553718511393, + "grad_norm": 0.7931581092579119, + "learning_rate": 2.334930164292294e-06, + "loss": 0.837, + "step": 3851 + }, + { + "epoch": 0.4631756147417784, + "grad_norm": 2.65476256966608, + "learning_rate": 2.334162167891612e-06, + "loss": 1.0368, + "step": 3852 + }, + { + "epoch": 0.4632958576324175, + "grad_norm": 2.0416773841213307, + "learning_rate": 2.333394120798795e-06, + "loss": 0.9849, + "step": 3853 + }, + { + "epoch": 0.4634161005230566, + "grad_norm": 2.497531244122614, + "learning_rate": 2.3326260231303545e-06, + "loss": 0.9546, + "step": 3854 + }, + { + "epoch": 0.46353634341369565, + "grad_norm": 1.6749170738303356, + "learning_rate": 2.331857875002811e-06, + "loss": 1.0983, + "step": 3855 + }, + { + "epoch": 0.46365658630433476, + "grad_norm": 1.902886785121263, + "learning_rate": 2.3310896765326916e-06, + "loss": 1.0016, + "step": 3856 + }, + { + "epoch": 0.46377682919497387, + "grad_norm": 1.8183547858847167, + "learning_rate": 2.330321427836531e-06, + "loss": 1.0792, + "step": 3857 + }, + { + "epoch": 0.4638970720856129, + "grad_norm": 2.0265146077775724, + "learning_rate": 2.3295531290308733e-06, + "loss": 1.0651, + "step": 3858 + }, + { + "epoch": 0.46401731497625204, + "grad_norm": 2.9446237097656907, + "learning_rate": 2.3287847802322678e-06, + "loss": 0.9888, + "step": 3859 + }, + { + "epoch": 0.4641375578668911, + "grad_norm": 2.174831217213018, + "learning_rate": 2.3280163815572723e-06, + "loss": 1.0776, + "step": 3860 + }, + { + "epoch": 0.4642578007575302, + "grad_norm": 2.1722995699840593, + "learning_rate": 2.3272479331224522e-06, + "loss": 1.0108, + "step": 3861 + }, + { + "epoch": 0.4643780436481693, + "grad_norm": 2.2703037752743898, + "learning_rate": 2.3264794350443817e-06, + "loss": 1.0146, + "step": 3862 + }, + { + "epoch": 0.46449828653880837, + "grad_norm": 1.9988336738597374, + "learning_rate": 2.3257108874396396e-06, + "loss": 1.0236, + "step": 3863 + }, + { + "epoch": 0.4646185294294475, + "grad_norm": 2.1936924070965658, + "learning_rate": 2.3249422904248152e-06, + "loss": 0.9733, + "step": 3864 + }, + { + "epoch": 0.4647387723200866, + "grad_norm": 1.9922532613035628, + "learning_rate": 2.324173644116504e-06, + "loss": 1.1119, + "step": 3865 + }, + { + "epoch": 0.46485901521072565, + "grad_norm": 1.9391119619450168, + "learning_rate": 2.3234049486313087e-06, + "loss": 1.0492, + "step": 3866 + }, + { + "epoch": 0.46497925810136476, + "grad_norm": 4.145481701736085, + "learning_rate": 2.322636204085839e-06, + "loss": 0.9926, + "step": 3867 + }, + { + "epoch": 0.46509950099200387, + "grad_norm": 2.220674996042716, + "learning_rate": 2.3218674105967143e-06, + "loss": 1.0269, + "step": 3868 + }, + { + "epoch": 0.4652197438826429, + "grad_norm": 1.6754554129427848, + "learning_rate": 2.3210985682805593e-06, + "loss": 1.0741, + "step": 3869 + }, + { + "epoch": 0.46533998677328203, + "grad_norm": 2.337316201883667, + "learning_rate": 2.320329677254007e-06, + "loss": 0.9158, + "step": 3870 + }, + { + "epoch": 0.46546022966392114, + "grad_norm": 2.501059879701121, + "learning_rate": 2.319560737633697e-06, + "loss": 0.9577, + "step": 3871 + }, + { + "epoch": 0.4655804725545602, + "grad_norm": 1.7661043714784244, + "learning_rate": 2.3187917495362775e-06, + "loss": 0.921, + "step": 3872 + }, + { + "epoch": 0.4657007154451993, + "grad_norm": 2.4129272556967356, + "learning_rate": 2.318022713078403e-06, + "loss": 1.0077, + "step": 3873 + }, + { + "epoch": 0.4658209583358384, + "grad_norm": 2.4050143153028265, + "learning_rate": 2.3172536283767354e-06, + "loss": 1.0912, + "step": 3874 + }, + { + "epoch": 0.4659412012264775, + "grad_norm": 2.070202518636804, + "learning_rate": 2.3164844955479447e-06, + "loss": 1.048, + "step": 3875 + }, + { + "epoch": 0.4660614441171166, + "grad_norm": 1.6671128232153274, + "learning_rate": 2.3157153147087082e-06, + "loss": 0.9412, + "step": 3876 + }, + { + "epoch": 0.46618168700775564, + "grad_norm": 1.966880698060933, + "learning_rate": 2.314946085975709e-06, + "loss": 1.0683, + "step": 3877 + }, + { + "epoch": 0.46630192989839475, + "grad_norm": 2.061745014146253, + "learning_rate": 2.3141768094656393e-06, + "loss": 1.0612, + "step": 3878 + }, + { + "epoch": 0.46642217278903386, + "grad_norm": 2.332061511790671, + "learning_rate": 2.3134074852951966e-06, + "loss": 1.0711, + "step": 3879 + }, + { + "epoch": 0.4665424156796729, + "grad_norm": 1.8000652354393043, + "learning_rate": 2.312638113581088e-06, + "loss": 1.0126, + "step": 3880 + }, + { + "epoch": 0.46666265857031203, + "grad_norm": 2.589707314880507, + "learning_rate": 2.311868694440027e-06, + "loss": 1.0129, + "step": 3881 + }, + { + "epoch": 0.46678290146095114, + "grad_norm": 0.8265990467320781, + "learning_rate": 2.3110992279887323e-06, + "loss": 0.8621, + "step": 3882 + }, + { + "epoch": 0.4669031443515902, + "grad_norm": 2.634775669313462, + "learning_rate": 2.310329714343932e-06, + "loss": 1.0895, + "step": 3883 + }, + { + "epoch": 0.4670233872422293, + "grad_norm": 2.4490069186703645, + "learning_rate": 2.309560153622361e-06, + "loss": 1.0596, + "step": 3884 + }, + { + "epoch": 0.4671436301328684, + "grad_norm": 2.6416399207300523, + "learning_rate": 2.3087905459407602e-06, + "loss": 0.9827, + "step": 3885 + }, + { + "epoch": 0.46726387302350747, + "grad_norm": 0.7960011796223704, + "learning_rate": 2.3080208914158795e-06, + "loss": 0.8673, + "step": 3886 + }, + { + "epoch": 0.4673841159141466, + "grad_norm": 2.266844953743313, + "learning_rate": 2.3072511901644753e-06, + "loss": 0.9635, + "step": 3887 + }, + { + "epoch": 0.4675043588047857, + "grad_norm": 2.2445430278635032, + "learning_rate": 2.306481442303309e-06, + "loss": 1.0437, + "step": 3888 + }, + { + "epoch": 0.46762460169542475, + "grad_norm": 1.8498603304127887, + "learning_rate": 2.3057116479491515e-06, + "loss": 0.9695, + "step": 3889 + }, + { + "epoch": 0.46774484458606386, + "grad_norm": 1.874226537754278, + "learning_rate": 2.30494180721878e-06, + "loss": 0.9958, + "step": 3890 + }, + { + "epoch": 0.4678650874767029, + "grad_norm": 2.0934489019447575, + "learning_rate": 2.3041719202289794e-06, + "loss": 1.1359, + "step": 3891 + }, + { + "epoch": 0.467985330367342, + "grad_norm": 2.1823931346480485, + "learning_rate": 2.30340198709654e-06, + "loss": 1.0446, + "step": 3892 + }, + { + "epoch": 0.46810557325798113, + "grad_norm": 2.75648514140342, + "learning_rate": 2.3026320079382605e-06, + "loss": 0.9783, + "step": 3893 + }, + { + "epoch": 0.4682258161486202, + "grad_norm": 5.112989727712004, + "learning_rate": 2.3018619828709454e-06, + "loss": 0.9992, + "step": 3894 + }, + { + "epoch": 0.4683460590392593, + "grad_norm": 2.0267238955615863, + "learning_rate": 2.3010919120114084e-06, + "loss": 1.059, + "step": 3895 + }, + { + "epoch": 0.4684663019298984, + "grad_norm": 2.2343353054548794, + "learning_rate": 2.3003217954764672e-06, + "loss": 0.8977, + "step": 3896 + }, + { + "epoch": 0.46858654482053747, + "grad_norm": 1.7630683144994117, + "learning_rate": 2.299551633382949e-06, + "loss": 1.0319, + "step": 3897 + }, + { + "epoch": 0.4687067877111766, + "grad_norm": 2.117683605894727, + "learning_rate": 2.2987814258476854e-06, + "loss": 1.0939, + "step": 3898 + }, + { + "epoch": 0.4688270306018157, + "grad_norm": 2.2352827669009017, + "learning_rate": 2.2980111729875177e-06, + "loss": 0.9157, + "step": 3899 + }, + { + "epoch": 0.46894727349245474, + "grad_norm": 1.7847353649851185, + "learning_rate": 2.2972408749192917e-06, + "loss": 1.0579, + "step": 3900 + }, + { + "epoch": 0.46906751638309385, + "grad_norm": 1.8519259711726754, + "learning_rate": 2.296470531759861e-06, + "loss": 0.905, + "step": 3901 + }, + { + "epoch": 0.46918775927373296, + "grad_norm": 3.9152593857545708, + "learning_rate": 2.2957001436260866e-06, + "loss": 1.0321, + "step": 3902 + }, + { + "epoch": 0.469308002164372, + "grad_norm": 1.7515026180978472, + "learning_rate": 2.294929710634836e-06, + "loss": 0.9675, + "step": 3903 + }, + { + "epoch": 0.46942824505501113, + "grad_norm": 1.992546976605443, + "learning_rate": 2.2941592329029823e-06, + "loss": 0.8503, + "step": 3904 + }, + { + "epoch": 0.46954848794565024, + "grad_norm": 1.9508620313021605, + "learning_rate": 2.2933887105474067e-06, + "loss": 1.0272, + "step": 3905 + }, + { + "epoch": 0.4696687308362893, + "grad_norm": 1.5622984946970258, + "learning_rate": 2.2926181436849974e-06, + "loss": 1.0515, + "step": 3906 + }, + { + "epoch": 0.4697889737269284, + "grad_norm": 1.7252105667301876, + "learning_rate": 2.2918475324326478e-06, + "loss": 0.966, + "step": 3907 + }, + { + "epoch": 0.46990921661756746, + "grad_norm": 2.637557556102875, + "learning_rate": 2.2910768769072603e-06, + "loss": 1.1531, + "step": 3908 + }, + { + "epoch": 0.47002945950820657, + "grad_norm": 1.9763091758057552, + "learning_rate": 2.2903061772257417e-06, + "loss": 0.9988, + "step": 3909 + }, + { + "epoch": 0.4701497023988457, + "grad_norm": 2.2343469364564656, + "learning_rate": 2.289535433505007e-06, + "loss": 1.0209, + "step": 3910 + }, + { + "epoch": 0.47026994528948474, + "grad_norm": 1.9798361465781362, + "learning_rate": 2.2887646458619767e-06, + "loss": 0.8705, + "step": 3911 + }, + { + "epoch": 0.47039018818012385, + "grad_norm": 1.997016887369897, + "learning_rate": 2.2879938144135797e-06, + "loss": 0.9993, + "step": 3912 + }, + { + "epoch": 0.47051043107076296, + "grad_norm": 1.9736380069745472, + "learning_rate": 2.2872229392767496e-06, + "loss": 0.988, + "step": 3913 + }, + { + "epoch": 0.470630673961402, + "grad_norm": 1.5356995686431039, + "learning_rate": 2.286452020568428e-06, + "loss": 0.9857, + "step": 3914 + }, + { + "epoch": 0.4707509168520411, + "grad_norm": 1.793353203671688, + "learning_rate": 2.2856810584055637e-06, + "loss": 0.9676, + "step": 3915 + }, + { + "epoch": 0.47087115974268023, + "grad_norm": 1.5438209980688724, + "learning_rate": 2.2849100529051085e-06, + "loss": 0.9157, + "step": 3916 + }, + { + "epoch": 0.4709914026333193, + "grad_norm": 2.4079968935734173, + "learning_rate": 2.284139004184026e-06, + "loss": 1.0393, + "step": 3917 + }, + { + "epoch": 0.4711116455239584, + "grad_norm": 3.62702227779727, + "learning_rate": 2.2833679123592814e-06, + "loss": 0.984, + "step": 3918 + }, + { + "epoch": 0.4712318884145975, + "grad_norm": 1.7414668350325568, + "learning_rate": 2.2825967775478508e-06, + "loss": 0.8734, + "step": 3919 + }, + { + "epoch": 0.47135213130523657, + "grad_norm": 3.511337310476847, + "learning_rate": 2.2818255998667135e-06, + "loss": 1.0724, + "step": 3920 + }, + { + "epoch": 0.4714723741958757, + "grad_norm": 1.6994682041211737, + "learning_rate": 2.2810543794328566e-06, + "loss": 1.0272, + "step": 3921 + }, + { + "epoch": 0.4715926170865148, + "grad_norm": 2.212286763233537, + "learning_rate": 2.2802831163632735e-06, + "loss": 1.0622, + "step": 3922 + }, + { + "epoch": 0.47171285997715384, + "grad_norm": 2.0586006989624175, + "learning_rate": 2.279511810774965e-06, + "loss": 0.9829, + "step": 3923 + }, + { + "epoch": 0.47183310286779295, + "grad_norm": 1.987053572701892, + "learning_rate": 2.2787404627849364e-06, + "loss": 0.9539, + "step": 3924 + }, + { + "epoch": 0.471953345758432, + "grad_norm": 1.7993551635708802, + "learning_rate": 2.277969072510202e-06, + "loss": 1.0313, + "step": 3925 + }, + { + "epoch": 0.4720735886490711, + "grad_norm": 1.764075620617656, + "learning_rate": 2.2771976400677803e-06, + "loss": 1.0511, + "step": 3926 + }, + { + "epoch": 0.47219383153971023, + "grad_norm": 1.8492814833810458, + "learning_rate": 2.2764261655746965e-06, + "loss": 1.0302, + "step": 3927 + }, + { + "epoch": 0.4723140744303493, + "grad_norm": 4.069926120921698, + "learning_rate": 2.2756546491479832e-06, + "loss": 0.9947, + "step": 3928 + }, + { + "epoch": 0.4724343173209884, + "grad_norm": 2.788099619603031, + "learning_rate": 2.274883090904679e-06, + "loss": 1.0438, + "step": 3929 + }, + { + "epoch": 0.4725545602116275, + "grad_norm": 2.434145257218588, + "learning_rate": 2.2741114909618283e-06, + "loss": 0.912, + "step": 3930 + }, + { + "epoch": 0.47267480310226656, + "grad_norm": 2.1829854974479037, + "learning_rate": 2.2733398494364828e-06, + "loss": 0.9553, + "step": 3931 + }, + { + "epoch": 0.47279504599290567, + "grad_norm": 2.0327222916968104, + "learning_rate": 2.272568166445699e-06, + "loss": 1.0833, + "step": 3932 + }, + { + "epoch": 0.4729152888835448, + "grad_norm": 1.8722000196532094, + "learning_rate": 2.271796442106541e-06, + "loss": 0.8833, + "step": 3933 + }, + { + "epoch": 0.47303553177418384, + "grad_norm": 0.8169084309491589, + "learning_rate": 2.271024676536079e-06, + "loss": 0.8035, + "step": 3934 + }, + { + "epoch": 0.47315577466482295, + "grad_norm": 2.0757301700439212, + "learning_rate": 2.2702528698513894e-06, + "loss": 0.9709, + "step": 3935 + }, + { + "epoch": 0.47327601755546206, + "grad_norm": 2.030293400800722, + "learning_rate": 2.269481022169554e-06, + "loss": 1.024, + "step": 3936 + }, + { + "epoch": 0.4733962604461011, + "grad_norm": 1.8659258933689402, + "learning_rate": 2.2687091336076614e-06, + "loss": 1.0429, + "step": 3937 + }, + { + "epoch": 0.4735165033367402, + "grad_norm": 1.993488560100536, + "learning_rate": 2.267937204282807e-06, + "loss": 1.0396, + "step": 3938 + }, + { + "epoch": 0.4736367462273793, + "grad_norm": 2.3644889330960903, + "learning_rate": 2.2671652343120926e-06, + "loss": 1.0264, + "step": 3939 + }, + { + "epoch": 0.4737569891180184, + "grad_norm": 2.3831626759854143, + "learning_rate": 2.2663932238126236e-06, + "loss": 1.0437, + "step": 3940 + }, + { + "epoch": 0.4738772320086575, + "grad_norm": 1.3893739881009883, + "learning_rate": 2.265621172901515e-06, + "loss": 1.0396, + "step": 3941 + }, + { + "epoch": 0.47399747489929656, + "grad_norm": 2.387190202515979, + "learning_rate": 2.2648490816958854e-06, + "loss": 0.951, + "step": 3942 + }, + { + "epoch": 0.47411771778993567, + "grad_norm": 2.5323873719667724, + "learning_rate": 2.264076950312861e-06, + "loss": 0.9702, + "step": 3943 + }, + { + "epoch": 0.4742379606805748, + "grad_norm": 2.0604545391781928, + "learning_rate": 2.2633047788695727e-06, + "loss": 1.0598, + "step": 3944 + }, + { + "epoch": 0.47435820357121383, + "grad_norm": 1.8239847886806455, + "learning_rate": 2.262532567483159e-06, + "loss": 0.8783, + "step": 3945 + }, + { + "epoch": 0.47447844646185294, + "grad_norm": 3.4333907100981786, + "learning_rate": 2.2617603162707635e-06, + "loss": 1.0412, + "step": 3946 + }, + { + "epoch": 0.47459868935249205, + "grad_norm": 1.6688453818504456, + "learning_rate": 2.2609880253495363e-06, + "loss": 1.0648, + "step": 3947 + }, + { + "epoch": 0.4747189322431311, + "grad_norm": 2.8673908878709256, + "learning_rate": 2.260215694836633e-06, + "loss": 1.1012, + "step": 3948 + }, + { + "epoch": 0.4748391751337702, + "grad_norm": 2.2791887141619456, + "learning_rate": 2.2594433248492157e-06, + "loss": 0.8846, + "step": 3949 + }, + { + "epoch": 0.47495941802440933, + "grad_norm": 1.8618886930097274, + "learning_rate": 2.2586709155044527e-06, + "loss": 1.0394, + "step": 3950 + }, + { + "epoch": 0.4750796609150484, + "grad_norm": 1.9261138814451377, + "learning_rate": 2.2578984669195167e-06, + "loss": 0.9994, + "step": 3951 + }, + { + "epoch": 0.4751999038056875, + "grad_norm": 1.9684721129953817, + "learning_rate": 2.2571259792115887e-06, + "loss": 0.9157, + "step": 3952 + }, + { + "epoch": 0.4753201466963266, + "grad_norm": 1.880762654290456, + "learning_rate": 2.2563534524978544e-06, + "loss": 1.03, + "step": 3953 + }, + { + "epoch": 0.47544038958696566, + "grad_norm": 2.189800469701962, + "learning_rate": 2.2555808868955052e-06, + "loss": 0.9436, + "step": 3954 + }, + { + "epoch": 0.47556063247760477, + "grad_norm": 2.8591018296127064, + "learning_rate": 2.254808282521738e-06, + "loss": 0.9768, + "step": 3955 + }, + { + "epoch": 0.4756808753682438, + "grad_norm": 1.818599304038661, + "learning_rate": 2.2540356394937573e-06, + "loss": 1.0495, + "step": 3956 + }, + { + "epoch": 0.47580111825888294, + "grad_norm": 2.1928551655685435, + "learning_rate": 2.253262957928772e-06, + "loss": 1.0734, + "step": 3957 + }, + { + "epoch": 0.47592136114952205, + "grad_norm": 1.6766427766578065, + "learning_rate": 2.2524902379439976e-06, + "loss": 0.9592, + "step": 3958 + }, + { + "epoch": 0.4760416040401611, + "grad_norm": 0.7477935282320807, + "learning_rate": 2.251717479656655e-06, + "loss": 0.8747, + "step": 3959 + }, + { + "epoch": 0.4761618469308002, + "grad_norm": 2.104252904755132, + "learning_rate": 2.2509446831839704e-06, + "loss": 1.0006, + "step": 3960 + }, + { + "epoch": 0.4762820898214393, + "grad_norm": 3.3139271091141222, + "learning_rate": 2.250171848643177e-06, + "loss": 1.0614, + "step": 3961 + }, + { + "epoch": 0.4764023327120784, + "grad_norm": 1.8849199459167088, + "learning_rate": 2.249398976151513e-06, + "loss": 1.096, + "step": 3962 + }, + { + "epoch": 0.4765225756027175, + "grad_norm": 2.0963543309434782, + "learning_rate": 2.248626065826223e-06, + "loss": 1.0195, + "step": 3963 + }, + { + "epoch": 0.4766428184933566, + "grad_norm": 0.7702931925924837, + "learning_rate": 2.2478531177845564e-06, + "loss": 0.8639, + "step": 3964 + }, + { + "epoch": 0.47676306138399566, + "grad_norm": 1.8822874231366313, + "learning_rate": 2.247080132143769e-06, + "loss": 1.0885, + "step": 3965 + }, + { + "epoch": 0.47688330427463477, + "grad_norm": 2.3997246584304825, + "learning_rate": 2.246307109021121e-06, + "loss": 0.9304, + "step": 3966 + }, + { + "epoch": 0.4770035471652739, + "grad_norm": 1.7293204120475807, + "learning_rate": 2.2455340485338817e-06, + "loss": 1.0629, + "step": 3967 + }, + { + "epoch": 0.47712379005591293, + "grad_norm": 3.0437239565508634, + "learning_rate": 2.244760950799322e-06, + "loss": 0.9144, + "step": 3968 + }, + { + "epoch": 0.47724403294655204, + "grad_norm": 2.1635299743522514, + "learning_rate": 2.2439878159347203e-06, + "loss": 0.9629, + "step": 3969 + }, + { + "epoch": 0.4773642758371911, + "grad_norm": 0.7367727621916085, + "learning_rate": 2.2432146440573616e-06, + "loss": 0.7908, + "step": 3970 + }, + { + "epoch": 0.4774845187278302, + "grad_norm": 2.0131799816734817, + "learning_rate": 2.242441435284534e-06, + "loss": 0.903, + "step": 3971 + }, + { + "epoch": 0.4776047616184693, + "grad_norm": 6.2896296085535095, + "learning_rate": 2.2416681897335337e-06, + "loss": 1.0902, + "step": 3972 + }, + { + "epoch": 0.4777250045091084, + "grad_norm": 1.9097817644797672, + "learning_rate": 2.240894907521661e-06, + "loss": 0.9037, + "step": 3973 + }, + { + "epoch": 0.4778452473997475, + "grad_norm": 1.7115580509024189, + "learning_rate": 2.240121588766223e-06, + "loss": 0.875, + "step": 3974 + }, + { + "epoch": 0.4779654902903866, + "grad_norm": 2.1173678951197146, + "learning_rate": 2.239348233584531e-06, + "loss": 0.9516, + "step": 3975 + }, + { + "epoch": 0.47808573318102565, + "grad_norm": 2.0226417655912434, + "learning_rate": 2.2385748420939013e-06, + "loss": 1.0442, + "step": 3976 + }, + { + "epoch": 0.47820597607166476, + "grad_norm": 2.603466997887266, + "learning_rate": 2.2378014144116583e-06, + "loss": 0.9583, + "step": 3977 + }, + { + "epoch": 0.4783262189623039, + "grad_norm": 1.9639494967367215, + "learning_rate": 2.23702795065513e-06, + "loss": 1.0314, + "step": 3978 + }, + { + "epoch": 0.47844646185294293, + "grad_norm": 0.9892587765296389, + "learning_rate": 2.2362544509416493e-06, + "loss": 0.9124, + "step": 3979 + }, + { + "epoch": 0.47856670474358204, + "grad_norm": 2.6256458759454104, + "learning_rate": 2.2354809153885572e-06, + "loss": 1.0651, + "step": 3980 + }, + { + "epoch": 0.47868694763422115, + "grad_norm": 1.7572213090048734, + "learning_rate": 2.234707344113197e-06, + "loss": 1.0696, + "step": 3981 + }, + { + "epoch": 0.4788071905248602, + "grad_norm": 1.7810596565137944, + "learning_rate": 2.233933737232919e-06, + "loss": 1.0154, + "step": 3982 + }, + { + "epoch": 0.4789274334154993, + "grad_norm": 3.6212018602810354, + "learning_rate": 2.2331600948650793e-06, + "loss": 1.0197, + "step": 3983 + }, + { + "epoch": 0.4790476763061384, + "grad_norm": 1.920536040043656, + "learning_rate": 2.2323864171270386e-06, + "loss": 1.0402, + "step": 3984 + }, + { + "epoch": 0.4791679191967775, + "grad_norm": 1.9133850767314249, + "learning_rate": 2.231612704136164e-06, + "loss": 0.9608, + "step": 3985 + }, + { + "epoch": 0.4792881620874166, + "grad_norm": 2.535087314132809, + "learning_rate": 2.2308389560098253e-06, + "loss": 0.9919, + "step": 3986 + }, + { + "epoch": 0.47940840497805565, + "grad_norm": 2.693801058044158, + "learning_rate": 2.2300651728654008e-06, + "loss": 1.0101, + "step": 3987 + }, + { + "epoch": 0.47952864786869476, + "grad_norm": 0.7406637511597989, + "learning_rate": 2.229291354820272e-06, + "loss": 0.8395, + "step": 3988 + }, + { + "epoch": 0.47964889075933387, + "grad_norm": 2.005837979894417, + "learning_rate": 2.228517501991828e-06, + "loss": 0.9982, + "step": 3989 + }, + { + "epoch": 0.4797691336499729, + "grad_norm": 0.8226988822128178, + "learning_rate": 2.22774361449746e-06, + "loss": 0.8494, + "step": 3990 + }, + { + "epoch": 0.47988937654061203, + "grad_norm": 2.7762342509063895, + "learning_rate": 2.2269696924545668e-06, + "loss": 0.9388, + "step": 3991 + }, + { + "epoch": 0.48000961943125114, + "grad_norm": 2.560373587022185, + "learning_rate": 2.2261957359805523e-06, + "loss": 1.0178, + "step": 3992 + }, + { + "epoch": 0.4801298623218902, + "grad_norm": 2.1486699758810532, + "learning_rate": 2.225421745192823e-06, + "loss": 0.9784, + "step": 3993 + }, + { + "epoch": 0.4802501052125293, + "grad_norm": 2.396992499742801, + "learning_rate": 2.2246477202087955e-06, + "loss": 1.0183, + "step": 3994 + }, + { + "epoch": 0.4803703481031684, + "grad_norm": 1.8036215800821922, + "learning_rate": 2.223873661145887e-06, + "loss": 1.0677, + "step": 3995 + }, + { + "epoch": 0.4804905909938075, + "grad_norm": 1.6623368606259685, + "learning_rate": 2.2230995681215226e-06, + "loss": 0.9504, + "step": 3996 + }, + { + "epoch": 0.4806108338844466, + "grad_norm": 2.06131490703521, + "learning_rate": 2.2223254412531305e-06, + "loss": 1.0159, + "step": 3997 + }, + { + "epoch": 0.4807310767750857, + "grad_norm": 2.1630665384288137, + "learning_rate": 2.221551280658146e-06, + "loss": 1.0632, + "step": 3998 + }, + { + "epoch": 0.48085131966572475, + "grad_norm": 1.6713655355115764, + "learning_rate": 2.2207770864540085e-06, + "loss": 0.978, + "step": 3999 + }, + { + "epoch": 0.48097156255636386, + "grad_norm": 2.1389516708385687, + "learning_rate": 2.220002858758162e-06, + "loss": 0.9655, + "step": 4000 + }, + { + "epoch": 0.481091805447003, + "grad_norm": 0.8813521873801461, + "learning_rate": 2.2192285976880573e-06, + "loss": 0.8466, + "step": 4001 + }, + { + "epoch": 0.48121204833764203, + "grad_norm": 1.6156197287733611, + "learning_rate": 2.2184543033611485e-06, + "loss": 1.0444, + "step": 4002 + }, + { + "epoch": 0.48133229122828114, + "grad_norm": 3.231059415940973, + "learning_rate": 2.2176799758948957e-06, + "loss": 1.0548, + "step": 4003 + }, + { + "epoch": 0.4814525341189202, + "grad_norm": 2.123969782009079, + "learning_rate": 2.2169056154067635e-06, + "loss": 0.9721, + "step": 4004 + }, + { + "epoch": 0.4815727770095593, + "grad_norm": 2.248357491199663, + "learning_rate": 2.216131222014222e-06, + "loss": 1.0612, + "step": 4005 + }, + { + "epoch": 0.4816930199001984, + "grad_norm": 2.0419665006155783, + "learning_rate": 2.2153567958347455e-06, + "loss": 1.0348, + "step": 4006 + }, + { + "epoch": 0.48181326279083747, + "grad_norm": 2.0209297340707493, + "learning_rate": 2.214582336985815e-06, + "loss": 1.0381, + "step": 4007 + }, + { + "epoch": 0.4819335056814766, + "grad_norm": 2.3514515796748134, + "learning_rate": 2.2138078455849142e-06, + "loss": 0.8979, + "step": 4008 + }, + { + "epoch": 0.4820537485721157, + "grad_norm": 2.7428188194992247, + "learning_rate": 2.2130333217495334e-06, + "loss": 1.0247, + "step": 4009 + }, + { + "epoch": 0.48217399146275475, + "grad_norm": 2.566751721033887, + "learning_rate": 2.2122587655971665e-06, + "loss": 0.9178, + "step": 4010 + }, + { + "epoch": 0.48229423435339386, + "grad_norm": 1.97592840465001, + "learning_rate": 2.211484177245314e-06, + "loss": 0.8768, + "step": 4011 + }, + { + "epoch": 0.48241447724403297, + "grad_norm": 2.001746130685984, + "learning_rate": 2.21070955681148e-06, + "loss": 0.9609, + "step": 4012 + }, + { + "epoch": 0.482534720134672, + "grad_norm": 2.2915496333900425, + "learning_rate": 2.209934904413174e-06, + "loss": 1.0169, + "step": 4013 + }, + { + "epoch": 0.48265496302531113, + "grad_norm": 2.2502218772843885, + "learning_rate": 2.2091602201679095e-06, + "loss": 0.954, + "step": 4014 + }, + { + "epoch": 0.48277520591595025, + "grad_norm": 2.361532778268321, + "learning_rate": 2.208385504193206e-06, + "loss": 1.0669, + "step": 4015 + }, + { + "epoch": 0.4828954488065893, + "grad_norm": 2.2019837582203152, + "learning_rate": 2.2076107566065873e-06, + "loss": 1.0477, + "step": 4016 + }, + { + "epoch": 0.4830156916972284, + "grad_norm": 2.4131573705470672, + "learning_rate": 2.2068359775255816e-06, + "loss": 0.9906, + "step": 4017 + }, + { + "epoch": 0.48313593458786747, + "grad_norm": 2.855988378250027, + "learning_rate": 2.206061167067723e-06, + "loss": 1.0266, + "step": 4018 + }, + { + "epoch": 0.4832561774785066, + "grad_norm": 2.049559369067001, + "learning_rate": 2.205286325350549e-06, + "loss": 1.0316, + "step": 4019 + }, + { + "epoch": 0.4833764203691457, + "grad_norm": 2.1369912149894055, + "learning_rate": 2.204511452491603e-06, + "loss": 0.9669, + "step": 4020 + }, + { + "epoch": 0.48349666325978474, + "grad_norm": 1.749706516179565, + "learning_rate": 2.2037365486084316e-06, + "loss": 0.9874, + "step": 4021 + }, + { + "epoch": 0.48361690615042385, + "grad_norm": 2.002533500095903, + "learning_rate": 2.2029616138185886e-06, + "loss": 1.0217, + "step": 4022 + }, + { + "epoch": 0.48373714904106296, + "grad_norm": 1.7844700402170872, + "learning_rate": 2.202186648239629e-06, + "loss": 1.0632, + "step": 4023 + }, + { + "epoch": 0.483857391931702, + "grad_norm": 1.7048499095554799, + "learning_rate": 2.201411651989117e-06, + "loss": 0.9507, + "step": 4024 + }, + { + "epoch": 0.48397763482234113, + "grad_norm": 1.9560544437250222, + "learning_rate": 2.2006366251846167e-06, + "loss": 1.0162, + "step": 4025 + }, + { + "epoch": 0.48409787771298024, + "grad_norm": 1.9113292252042167, + "learning_rate": 2.1998615679436997e-06, + "loss": 0.9918, + "step": 4026 + }, + { + "epoch": 0.4842181206036193, + "grad_norm": 2.5379032246842175, + "learning_rate": 2.199086480383942e-06, + "loss": 1.0113, + "step": 4027 + }, + { + "epoch": 0.4843383634942584, + "grad_norm": 2.686652560027441, + "learning_rate": 2.1983113626229234e-06, + "loss": 0.9123, + "step": 4028 + }, + { + "epoch": 0.4844586063848975, + "grad_norm": 2.1855336752212544, + "learning_rate": 2.1975362147782293e-06, + "loss": 1.0233, + "step": 4029 + }, + { + "epoch": 0.48457884927553657, + "grad_norm": 0.6947427193835677, + "learning_rate": 2.196761036967448e-06, + "loss": 0.779, + "step": 4030 + }, + { + "epoch": 0.4846990921661757, + "grad_norm": 1.872272032852716, + "learning_rate": 2.1959858293081743e-06, + "loss": 1.0127, + "step": 4031 + }, + { + "epoch": 0.4848193350568148, + "grad_norm": 1.7129094156519917, + "learning_rate": 2.1952105919180056e-06, + "loss": 0.9956, + "step": 4032 + }, + { + "epoch": 0.48493957794745385, + "grad_norm": 2.247793811437738, + "learning_rate": 2.1944353249145456e-06, + "loss": 0.9205, + "step": 4033 + }, + { + "epoch": 0.48505982083809296, + "grad_norm": 1.6758348860971373, + "learning_rate": 2.193660028415401e-06, + "loss": 0.9821, + "step": 4034 + }, + { + "epoch": 0.485180063728732, + "grad_norm": 2.381540728265415, + "learning_rate": 2.1928847025381852e-06, + "loss": 1.0573, + "step": 4035 + }, + { + "epoch": 0.4853003066193711, + "grad_norm": 1.8099144218825856, + "learning_rate": 2.192109347400512e-06, + "loss": 1.0767, + "step": 4036 + }, + { + "epoch": 0.48542054951001024, + "grad_norm": 1.8879689246747997, + "learning_rate": 2.191333963120004e-06, + "loss": 1.0272, + "step": 4037 + }, + { + "epoch": 0.4855407924006493, + "grad_norm": 3.2899373175496462, + "learning_rate": 2.190558549814286e-06, + "loss": 0.9375, + "step": 4038 + }, + { + "epoch": 0.4856610352912884, + "grad_norm": 1.976644401210463, + "learning_rate": 2.1897831076009872e-06, + "loss": 1.0342, + "step": 4039 + }, + { + "epoch": 0.4857812781819275, + "grad_norm": 2.065303977224918, + "learning_rate": 2.1890076365977426e-06, + "loss": 1.0341, + "step": 4040 + }, + { + "epoch": 0.48590152107256657, + "grad_norm": 0.8634813516963324, + "learning_rate": 2.188232136922189e-06, + "loss": 0.7659, + "step": 4041 + }, + { + "epoch": 0.4860217639632057, + "grad_norm": 1.9858329403271635, + "learning_rate": 2.187456608691971e-06, + "loss": 0.999, + "step": 4042 + }, + { + "epoch": 0.4861420068538448, + "grad_norm": 2.087905127222495, + "learning_rate": 2.1866810520247334e-06, + "loss": 1.115, + "step": 4043 + }, + { + "epoch": 0.48626224974448384, + "grad_norm": 1.9244562198439192, + "learning_rate": 2.185905467038129e-06, + "loss": 0.8865, + "step": 4044 + }, + { + "epoch": 0.48638249263512295, + "grad_norm": 1.7344000874672407, + "learning_rate": 2.1851298538498127e-06, + "loss": 1.0145, + "step": 4045 + }, + { + "epoch": 0.48650273552576206, + "grad_norm": 2.0713715921279214, + "learning_rate": 2.184354212577446e-06, + "loss": 1.0369, + "step": 4046 + }, + { + "epoch": 0.4866229784164011, + "grad_norm": 3.8271737298618835, + "learning_rate": 2.1835785433386907e-06, + "loss": 0.8607, + "step": 4047 + }, + { + "epoch": 0.48674322130704023, + "grad_norm": 2.0530530571489742, + "learning_rate": 2.182802846251216e-06, + "loss": 0.8891, + "step": 4048 + }, + { + "epoch": 0.4868634641976793, + "grad_norm": 2.321222608463215, + "learning_rate": 2.182027121432696e-06, + "loss": 0.9594, + "step": 4049 + }, + { + "epoch": 0.4869837070883184, + "grad_norm": 2.0162000205555457, + "learning_rate": 2.1812513690008054e-06, + "loss": 1.0564, + "step": 4050 + }, + { + "epoch": 0.4871039499789575, + "grad_norm": 2.425510360654754, + "learning_rate": 2.180475589073227e-06, + "loss": 1.0351, + "step": 4051 + }, + { + "epoch": 0.48722419286959656, + "grad_norm": 1.578132553837893, + "learning_rate": 2.1796997817676456e-06, + "loss": 0.9721, + "step": 4052 + }, + { + "epoch": 0.4873444357602357, + "grad_norm": 1.6813563373503528, + "learning_rate": 2.1789239472017494e-06, + "loss": 0.9109, + "step": 4053 + }, + { + "epoch": 0.4874646786508748, + "grad_norm": 2.203258970259024, + "learning_rate": 2.1781480854932326e-06, + "loss": 0.967, + "step": 4054 + }, + { + "epoch": 0.48758492154151384, + "grad_norm": 1.9389716681170828, + "learning_rate": 2.1773721967597933e-06, + "loss": 1.0316, + "step": 4055 + }, + { + "epoch": 0.48770516443215295, + "grad_norm": 0.851065359326647, + "learning_rate": 2.1765962811191322e-06, + "loss": 0.8111, + "step": 4056 + }, + { + "epoch": 0.48782540732279206, + "grad_norm": 0.8217569984460653, + "learning_rate": 2.1758203386889566e-06, + "loss": 0.857, + "step": 4057 + }, + { + "epoch": 0.4879456502134311, + "grad_norm": 2.0249486751880887, + "learning_rate": 2.1750443695869746e-06, + "loss": 1.0816, + "step": 4058 + }, + { + "epoch": 0.4880658931040702, + "grad_norm": 2.265855343388961, + "learning_rate": 2.174268373930901e-06, + "loss": 1.0934, + "step": 4059 + }, + { + "epoch": 0.48818613599470934, + "grad_norm": 2.1908628500595366, + "learning_rate": 2.1734923518384537e-06, + "loss": 1.0348, + "step": 4060 + }, + { + "epoch": 0.4883063788853484, + "grad_norm": 1.871356220989195, + "learning_rate": 2.1727163034273547e-06, + "loss": 1.0629, + "step": 4061 + }, + { + "epoch": 0.4884266217759875, + "grad_norm": 2.2498374456389767, + "learning_rate": 2.17194022881533e-06, + "loss": 1.0331, + "step": 4062 + }, + { + "epoch": 0.4885468646666266, + "grad_norm": 1.693188369370565, + "learning_rate": 2.1711641281201092e-06, + "loss": 0.9165, + "step": 4063 + }, + { + "epoch": 0.48866710755726567, + "grad_norm": 2.2014357130310205, + "learning_rate": 2.1703880014594264e-06, + "loss": 1.0306, + "step": 4064 + }, + { + "epoch": 0.4887873504479048, + "grad_norm": 1.8347581037692473, + "learning_rate": 2.1696118489510182e-06, + "loss": 0.9772, + "step": 4065 + }, + { + "epoch": 0.48890759333854383, + "grad_norm": 1.8284386259840604, + "learning_rate": 2.1688356707126286e-06, + "loss": 0.9603, + "step": 4066 + }, + { + "epoch": 0.48902783622918294, + "grad_norm": 2.1982707119637, + "learning_rate": 2.168059466862001e-06, + "loss": 0.9349, + "step": 4067 + }, + { + "epoch": 0.48914807911982205, + "grad_norm": 2.2295318613352957, + "learning_rate": 2.167283237516887e-06, + "loss": 1.0546, + "step": 4068 + }, + { + "epoch": 0.4892683220104611, + "grad_norm": 2.152289914325014, + "learning_rate": 2.1665069827950383e-06, + "loss": 0.9813, + "step": 4069 + }, + { + "epoch": 0.4893885649011002, + "grad_norm": 1.7782767393669867, + "learning_rate": 2.1657307028142126e-06, + "loss": 1.1093, + "step": 4070 + }, + { + "epoch": 0.48950880779173933, + "grad_norm": 1.8543321199926868, + "learning_rate": 2.164954397692171e-06, + "loss": 0.9127, + "step": 4071 + }, + { + "epoch": 0.4896290506823784, + "grad_norm": 1.105423404046635, + "learning_rate": 2.164178067546678e-06, + "loss": 1.011, + "step": 4072 + }, + { + "epoch": 0.4897492935730175, + "grad_norm": 1.7412382541980551, + "learning_rate": 2.163401712495504e-06, + "loss": 1.1435, + "step": 4073 + }, + { + "epoch": 0.4898695364636566, + "grad_norm": 1.7184204479203244, + "learning_rate": 2.1626253326564194e-06, + "loss": 1.0284, + "step": 4074 + }, + { + "epoch": 0.48998977935429566, + "grad_norm": 1.9431273662151507, + "learning_rate": 2.161848928147201e-06, + "loss": 1.0077, + "step": 4075 + }, + { + "epoch": 0.4901100222449348, + "grad_norm": 2.0138823790463287, + "learning_rate": 2.161072499085629e-06, + "loss": 1.043, + "step": 4076 + }, + { + "epoch": 0.4902302651355739, + "grad_norm": 1.9360607092615378, + "learning_rate": 2.160296045589487e-06, + "loss": 1.067, + "step": 4077 + }, + { + "epoch": 0.49035050802621294, + "grad_norm": 1.9162339952252145, + "learning_rate": 2.159519567776562e-06, + "loss": 0.9364, + "step": 4078 + }, + { + "epoch": 0.49047075091685205, + "grad_norm": 2.818373756081206, + "learning_rate": 2.1587430657646463e-06, + "loss": 0.9456, + "step": 4079 + }, + { + "epoch": 0.4905909938074911, + "grad_norm": 1.8796801331487498, + "learning_rate": 2.157966539671533e-06, + "loss": 1.017, + "step": 4080 + }, + { + "epoch": 0.4907112366981302, + "grad_norm": 2.0715037247690815, + "learning_rate": 2.157189989615021e-06, + "loss": 0.908, + "step": 4081 + }, + { + "epoch": 0.4908314795887693, + "grad_norm": 1.9082959561525854, + "learning_rate": 2.156413415712913e-06, + "loss": 0.9847, + "step": 4082 + }, + { + "epoch": 0.4909517224794084, + "grad_norm": 2.0399473177895278, + "learning_rate": 2.155636818083014e-06, + "loss": 1.0212, + "step": 4083 + }, + { + "epoch": 0.4910719653700475, + "grad_norm": 2.054159698577679, + "learning_rate": 2.154860196843134e-06, + "loss": 1.0834, + "step": 4084 + }, + { + "epoch": 0.4911922082606866, + "grad_norm": 2.071344543030537, + "learning_rate": 2.154083552111085e-06, + "loss": 1.0043, + "step": 4085 + }, + { + "epoch": 0.49131245115132566, + "grad_norm": 1.983083950787657, + "learning_rate": 2.1533068840046834e-06, + "loss": 1.0559, + "step": 4086 + }, + { + "epoch": 0.49143269404196477, + "grad_norm": 2.67043457466494, + "learning_rate": 2.152530192641749e-06, + "loss": 0.8562, + "step": 4087 + }, + { + "epoch": 0.4915529369326039, + "grad_norm": 1.8463082749474764, + "learning_rate": 2.1517534781401068e-06, + "loss": 0.9618, + "step": 4088 + }, + { + "epoch": 0.49167317982324293, + "grad_norm": 2.0725044594637354, + "learning_rate": 2.150976740617581e-06, + "loss": 0.9288, + "step": 4089 + }, + { + "epoch": 0.49179342271388204, + "grad_norm": 1.9424425885959808, + "learning_rate": 2.150199980192006e-06, + "loss": 0.9528, + "step": 4090 + }, + { + "epoch": 0.49191366560452116, + "grad_norm": 2.004251729173879, + "learning_rate": 2.1494231969812114e-06, + "loss": 1.0455, + "step": 4091 + }, + { + "epoch": 0.4920339084951602, + "grad_norm": 2.1756417522536124, + "learning_rate": 2.1486463911030372e-06, + "loss": 1.0499, + "step": 4092 + }, + { + "epoch": 0.4921541513857993, + "grad_norm": 4.246863610348505, + "learning_rate": 2.147869562675324e-06, + "loss": 0.9823, + "step": 4093 + }, + { + "epoch": 0.49227439427643843, + "grad_norm": 2.338658613808026, + "learning_rate": 2.147092711815915e-06, + "loss": 0.9638, + "step": 4094 + }, + { + "epoch": 0.4923946371670775, + "grad_norm": 2.3657850889110916, + "learning_rate": 2.1463158386426593e-06, + "loss": 1.1012, + "step": 4095 + }, + { + "epoch": 0.4925148800577166, + "grad_norm": 4.2762379510258715, + "learning_rate": 2.145538943273407e-06, + "loss": 1.0124, + "step": 4096 + }, + { + "epoch": 0.49263512294835565, + "grad_norm": 1.867175816455503, + "learning_rate": 2.144762025826013e-06, + "loss": 0.9513, + "step": 4097 + }, + { + "epoch": 0.49275536583899476, + "grad_norm": 2.3385899007080844, + "learning_rate": 2.143985086418334e-06, + "loss": 1.1072, + "step": 4098 + }, + { + "epoch": 0.4928756087296339, + "grad_norm": 1.4982171589781446, + "learning_rate": 2.1432081251682324e-06, + "loss": 1.0042, + "step": 4099 + }, + { + "epoch": 0.49299585162027293, + "grad_norm": 1.7595426056135266, + "learning_rate": 2.142431142193572e-06, + "loss": 1.1079, + "step": 4100 + }, + { + "epoch": 0.49311609451091204, + "grad_norm": 2.136252957470913, + "learning_rate": 2.1416541376122207e-06, + "loss": 0.9552, + "step": 4101 + }, + { + "epoch": 0.49323633740155115, + "grad_norm": 2.342260981266608, + "learning_rate": 2.1408771115420496e-06, + "loss": 0.9683, + "step": 4102 + }, + { + "epoch": 0.4933565802921902, + "grad_norm": 1.9661403495931047, + "learning_rate": 2.140100064100932e-06, + "loss": 0.8879, + "step": 4103 + }, + { + "epoch": 0.4934768231828293, + "grad_norm": 2.300676793575337, + "learning_rate": 2.139322995406746e-06, + "loss": 0.9953, + "step": 4104 + }, + { + "epoch": 0.4935970660734684, + "grad_norm": 3.5752754692171806, + "learning_rate": 2.1385459055773727e-06, + "loss": 1.0295, + "step": 4105 + }, + { + "epoch": 0.4937173089641075, + "grad_norm": 2.050945398936536, + "learning_rate": 2.137768794730696e-06, + "loss": 0.9706, + "step": 4106 + }, + { + "epoch": 0.4938375518547466, + "grad_norm": 3.6679838300076058, + "learning_rate": 2.1369916629846026e-06, + "loss": 1.0399, + "step": 4107 + }, + { + "epoch": 0.4939577947453857, + "grad_norm": 1.8407056858790838, + "learning_rate": 2.136214510456983e-06, + "loss": 0.9812, + "step": 4108 + }, + { + "epoch": 0.49407803763602476, + "grad_norm": 0.9533173413924968, + "learning_rate": 2.1354373372657296e-06, + "loss": 0.8693, + "step": 4109 + }, + { + "epoch": 0.49419828052666387, + "grad_norm": 1.6143752261348836, + "learning_rate": 2.1346601435287404e-06, + "loss": 0.9459, + "step": 4110 + }, + { + "epoch": 0.494318523417303, + "grad_norm": 1.9814249773707018, + "learning_rate": 2.1338829293639144e-06, + "loss": 1.0433, + "step": 4111 + }, + { + "epoch": 0.49443876630794203, + "grad_norm": 2.9112928385549415, + "learning_rate": 2.1331056948891547e-06, + "loss": 1.067, + "step": 4112 + }, + { + "epoch": 0.49455900919858115, + "grad_norm": 2.308913671849854, + "learning_rate": 2.1323284402223666e-06, + "loss": 0.9963, + "step": 4113 + }, + { + "epoch": 0.4946792520892202, + "grad_norm": 2.117063962140422, + "learning_rate": 2.1315511654814597e-06, + "loss": 1.1163, + "step": 4114 + }, + { + "epoch": 0.4947994949798593, + "grad_norm": 1.9624141285569259, + "learning_rate": 2.1307738707843456e-06, + "loss": 1.0217, + "step": 4115 + }, + { + "epoch": 0.4949197378704984, + "grad_norm": 2.111184481411035, + "learning_rate": 2.1299965562489385e-06, + "loss": 0.9259, + "step": 4116 + }, + { + "epoch": 0.4950399807611375, + "grad_norm": 1.4429347093632579, + "learning_rate": 2.129219221993158e-06, + "loss": 1.0264, + "step": 4117 + }, + { + "epoch": 0.4951602236517766, + "grad_norm": 0.7988442790069935, + "learning_rate": 2.128441868134924e-06, + "loss": 0.837, + "step": 4118 + }, + { + "epoch": 0.4952804665424157, + "grad_norm": 2.0920766719896178, + "learning_rate": 2.1276644947921606e-06, + "loss": 1.0662, + "step": 4119 + }, + { + "epoch": 0.49540070943305475, + "grad_norm": 1.8188755110073467, + "learning_rate": 2.126887102082795e-06, + "loss": 1.0661, + "step": 4120 + }, + { + "epoch": 0.49552095232369386, + "grad_norm": 2.0908150670619627, + "learning_rate": 2.126109690124757e-06, + "loss": 0.9419, + "step": 4121 + }, + { + "epoch": 0.495641195214333, + "grad_norm": 1.765180751309364, + "learning_rate": 2.1253322590359786e-06, + "loss": 0.9499, + "step": 4122 + }, + { + "epoch": 0.49576143810497203, + "grad_norm": 1.96105409556722, + "learning_rate": 2.124554808934397e-06, + "loss": 0.9761, + "step": 4123 + }, + { + "epoch": 0.49588168099561114, + "grad_norm": 2.965975458723981, + "learning_rate": 2.1237773399379496e-06, + "loss": 0.9662, + "step": 4124 + }, + { + "epoch": 0.49600192388625025, + "grad_norm": 2.016553679629771, + "learning_rate": 2.122999852164578e-06, + "loss": 1.1087, + "step": 4125 + }, + { + "epoch": 0.4961221667768893, + "grad_norm": 20.37430201546025, + "learning_rate": 2.122222345732227e-06, + "loss": 0.8241, + "step": 4126 + }, + { + "epoch": 0.4962424096675284, + "grad_norm": 2.572692696608291, + "learning_rate": 2.121444820758843e-06, + "loss": 1.069, + "step": 4127 + }, + { + "epoch": 0.49636265255816747, + "grad_norm": 3.446251566059475, + "learning_rate": 2.120667277362376e-06, + "loss": 1.024, + "step": 4128 + }, + { + "epoch": 0.4964828954488066, + "grad_norm": 2.106404904971441, + "learning_rate": 2.1198897156607796e-06, + "loss": 1.0887, + "step": 4129 + }, + { + "epoch": 0.4966031383394457, + "grad_norm": 2.287051101580692, + "learning_rate": 2.1191121357720085e-06, + "loss": 0.9715, + "step": 4130 + }, + { + "epoch": 0.49672338123008475, + "grad_norm": 4.983939316222468, + "learning_rate": 2.1183345378140206e-06, + "loss": 0.9818, + "step": 4131 + }, + { + "epoch": 0.49684362412072386, + "grad_norm": 0.8673797944086622, + "learning_rate": 2.1175569219047783e-06, + "loss": 0.8578, + "step": 4132 + }, + { + "epoch": 0.49696386701136297, + "grad_norm": 1.6958031713379356, + "learning_rate": 2.1167792881622437e-06, + "loss": 0.9725, + "step": 4133 + }, + { + "epoch": 0.497084109902002, + "grad_norm": 1.6072515163322507, + "learning_rate": 2.116001636704384e-06, + "loss": 1.0464, + "step": 4134 + }, + { + "epoch": 0.49720435279264114, + "grad_norm": 2.0180741445419326, + "learning_rate": 2.1152239676491685e-06, + "loss": 1.036, + "step": 4135 + }, + { + "epoch": 0.49732459568328025, + "grad_norm": 3.7856948831477446, + "learning_rate": 2.114446281114569e-06, + "loss": 0.9705, + "step": 4136 + }, + { + "epoch": 0.4974448385739193, + "grad_norm": 2.243574398044185, + "learning_rate": 2.1136685772185587e-06, + "loss": 0.996, + "step": 4137 + }, + { + "epoch": 0.4975650814645584, + "grad_norm": 1.6809657711013934, + "learning_rate": 2.1128908560791163e-06, + "loss": 1.0132, + "step": 4138 + }, + { + "epoch": 0.4976853243551975, + "grad_norm": 1.7478989522564496, + "learning_rate": 2.1121131178142203e-06, + "loss": 1.0184, + "step": 4139 + }, + { + "epoch": 0.4978055672458366, + "grad_norm": 1.8204178022452553, + "learning_rate": 2.1113353625418544e-06, + "loss": 1.0597, + "step": 4140 + }, + { + "epoch": 0.4979258101364757, + "grad_norm": 1.7496781734457123, + "learning_rate": 2.1105575903800017e-06, + "loss": 1.0273, + "step": 4141 + }, + { + "epoch": 0.4980460530271148, + "grad_norm": 2.376482651351634, + "learning_rate": 2.1097798014466502e-06, + "loss": 1.0896, + "step": 4142 + }, + { + "epoch": 0.49816629591775385, + "grad_norm": 2.2858929351405095, + "learning_rate": 2.109001995859791e-06, + "loss": 0.8253, + "step": 4143 + }, + { + "epoch": 0.49828653880839296, + "grad_norm": 0.8048135232850033, + "learning_rate": 2.108224173737415e-06, + "loss": 0.8384, + "step": 4144 + }, + { + "epoch": 0.498406781699032, + "grad_norm": 1.991259549507763, + "learning_rate": 2.1074463351975183e-06, + "loss": 0.9993, + "step": 4145 + }, + { + "epoch": 0.49852702458967113, + "grad_norm": 1.7789029502921816, + "learning_rate": 2.106668480358098e-06, + "loss": 0.9519, + "step": 4146 + }, + { + "epoch": 0.49864726748031024, + "grad_norm": 4.004123231544399, + "learning_rate": 2.105890609337154e-06, + "loss": 0.9448, + "step": 4147 + }, + { + "epoch": 0.4987675103709493, + "grad_norm": 0.704585233317708, + "learning_rate": 2.1051127222526883e-06, + "loss": 0.8757, + "step": 4148 + }, + { + "epoch": 0.4988877532615884, + "grad_norm": 1.7913550468144384, + "learning_rate": 2.1043348192227067e-06, + "loss": 1.0448, + "step": 4149 + }, + { + "epoch": 0.4990079961522275, + "grad_norm": 1.9568076243904715, + "learning_rate": 2.1035569003652156e-06, + "loss": 0.8557, + "step": 4150 + }, + { + "epoch": 0.4991282390428666, + "grad_norm": 2.1929849795946024, + "learning_rate": 2.1027789657982255e-06, + "loss": 1.0556, + "step": 4151 + }, + { + "epoch": 0.4992484819335057, + "grad_norm": 1.974531675898512, + "learning_rate": 2.1020010156397482e-06, + "loss": 1.0084, + "step": 4152 + }, + { + "epoch": 0.4993687248241448, + "grad_norm": 1.5527701511699992, + "learning_rate": 2.101223050007797e-06, + "loss": 1.012, + "step": 4153 + }, + { + "epoch": 0.49948896771478385, + "grad_norm": 0.832744024442536, + "learning_rate": 2.1004450690203904e-06, + "loss": 0.776, + "step": 4154 + }, + { + "epoch": 0.49960921060542296, + "grad_norm": 0.860687605908466, + "learning_rate": 2.099667072795546e-06, + "loss": 0.8701, + "step": 4155 + }, + { + "epoch": 0.49972945349606207, + "grad_norm": 1.8001135790276963, + "learning_rate": 2.0988890614512864e-06, + "loss": 1.035, + "step": 4156 + }, + { + "epoch": 0.4998496963867011, + "grad_norm": 2.3321743766275382, + "learning_rate": 2.098111035105635e-06, + "loss": 1.0786, + "step": 4157 + }, + { + "epoch": 0.49996993927734024, + "grad_norm": 1.8934448902862924, + "learning_rate": 2.0973329938766176e-06, + "loss": 0.9673, + "step": 4158 + }, + { + "epoch": 0.5000901821679793, + "grad_norm": 2.1252243821103707, + "learning_rate": 2.0965549378822618e-06, + "loss": 1.0259, + "step": 4159 + }, + { + "epoch": 0.5002104250586185, + "grad_norm": 2.096249696643969, + "learning_rate": 2.095776867240599e-06, + "loss": 1.078, + "step": 4160 + }, + { + "epoch": 0.5003306679492575, + "grad_norm": 2.579317129385774, + "learning_rate": 2.094998782069661e-06, + "loss": 1.0573, + "step": 4161 + }, + { + "epoch": 0.5004509108398966, + "grad_norm": 1.6090592750551078, + "learning_rate": 2.0942206824874845e-06, + "loss": 0.9911, + "step": 4162 + }, + { + "epoch": 0.5005711537305357, + "grad_norm": 2.456487986753638, + "learning_rate": 2.093442568612105e-06, + "loss": 1.0306, + "step": 4163 + }, + { + "epoch": 0.5006913966211748, + "grad_norm": 1.606187051808708, + "learning_rate": 2.0926644405615613e-06, + "loss": 1.0898, + "step": 4164 + }, + { + "epoch": 0.5008116395118138, + "grad_norm": 1.9913991527647903, + "learning_rate": 2.091886298453897e-06, + "loss": 1.0526, + "step": 4165 + }, + { + "epoch": 0.500931882402453, + "grad_norm": 2.3456505762419035, + "learning_rate": 2.091108142407153e-06, + "loss": 0.9635, + "step": 4166 + }, + { + "epoch": 0.5010521252930921, + "grad_norm": 0.8537779520577835, + "learning_rate": 2.090329972539377e-06, + "loss": 0.8611, + "step": 4167 + }, + { + "epoch": 0.5011723681837311, + "grad_norm": 1.980398744929684, + "learning_rate": 2.089551788968616e-06, + "loss": 0.9197, + "step": 4168 + }, + { + "epoch": 0.5012926110743702, + "grad_norm": 0.8402403493870112, + "learning_rate": 2.08877359181292e-06, + "loss": 0.8451, + "step": 4169 + }, + { + "epoch": 0.5014128539650093, + "grad_norm": 2.205119643652531, + "learning_rate": 2.0879953811903396e-06, + "loss": 1.0929, + "step": 4170 + }, + { + "epoch": 0.5015330968556484, + "grad_norm": 1.7928542555610452, + "learning_rate": 2.08721715721893e-06, + "loss": 1.0188, + "step": 4171 + }, + { + "epoch": 0.5016533397462875, + "grad_norm": 1.8279233641973118, + "learning_rate": 2.0864389200167477e-06, + "loss": 1.0051, + "step": 4172 + }, + { + "epoch": 0.5017735826369266, + "grad_norm": 1.8663584891875284, + "learning_rate": 2.0856606697018504e-06, + "loss": 1.0246, + "step": 4173 + }, + { + "epoch": 0.5018938255275657, + "grad_norm": 2.0907217872921087, + "learning_rate": 2.084882406392297e-06, + "loss": 0.9723, + "step": 4174 + }, + { + "epoch": 0.5020140684182047, + "grad_norm": 3.0691173450140057, + "learning_rate": 2.0841041302061496e-06, + "loss": 0.9477, + "step": 4175 + }, + { + "epoch": 0.5021343113088439, + "grad_norm": 1.9732109271996785, + "learning_rate": 2.083325841261473e-06, + "loss": 0.9936, + "step": 4176 + }, + { + "epoch": 0.502254554199483, + "grad_norm": 2.0135526427555117, + "learning_rate": 2.0825475396763322e-06, + "loss": 0.8977, + "step": 4177 + }, + { + "epoch": 0.502374797090122, + "grad_norm": 1.42930964912238, + "learning_rate": 2.081769225568796e-06, + "loss": 0.8952, + "step": 4178 + }, + { + "epoch": 0.5024950399807612, + "grad_norm": 1.716679452953435, + "learning_rate": 2.0809908990569327e-06, + "loss": 0.9999, + "step": 4179 + }, + { + "epoch": 0.5026152828714002, + "grad_norm": 1.9912116559126953, + "learning_rate": 2.0802125602588146e-06, + "loss": 1.0245, + "step": 4180 + }, + { + "epoch": 0.5027355257620393, + "grad_norm": 1.830354749033356, + "learning_rate": 2.0794342092925146e-06, + "loss": 0.9019, + "step": 4181 + }, + { + "epoch": 0.5028557686526784, + "grad_norm": 2.7748693933767536, + "learning_rate": 2.078655846276108e-06, + "loss": 0.9143, + "step": 4182 + }, + { + "epoch": 0.5029760115433175, + "grad_norm": 2.2108130537939394, + "learning_rate": 2.0778774713276727e-06, + "loss": 0.9257, + "step": 4183 + }, + { + "epoch": 0.5030962544339566, + "grad_norm": 3.349931938277085, + "learning_rate": 2.077099084565287e-06, + "loss": 0.9108, + "step": 4184 + }, + { + "epoch": 0.5032164973245957, + "grad_norm": 2.6714418294357603, + "learning_rate": 2.0763206861070313e-06, + "loss": 0.8881, + "step": 4185 + }, + { + "epoch": 0.5033367402152348, + "grad_norm": 1.820850996663588, + "learning_rate": 2.0755422760709876e-06, + "loss": 0.9917, + "step": 4186 + }, + { + "epoch": 0.5034569831058738, + "grad_norm": 1.898868841453953, + "learning_rate": 2.0747638545752417e-06, + "loss": 1.0054, + "step": 4187 + }, + { + "epoch": 0.503577225996513, + "grad_norm": 1.96790864920075, + "learning_rate": 2.073985421737878e-06, + "loss": 1.0689, + "step": 4188 + }, + { + "epoch": 0.5036974688871521, + "grad_norm": 2.5646357126307384, + "learning_rate": 2.0732069776769844e-06, + "loss": 0.9764, + "step": 4189 + }, + { + "epoch": 0.5038177117777911, + "grad_norm": 2.817622796321586, + "learning_rate": 2.072428522510651e-06, + "loss": 0.9662, + "step": 4190 + }, + { + "epoch": 0.5039379546684303, + "grad_norm": 2.170649120791018, + "learning_rate": 2.071650056356968e-06, + "loss": 1.0032, + "step": 4191 + }, + { + "epoch": 0.5040581975590693, + "grad_norm": 2.0433133420023313, + "learning_rate": 2.070871579334028e-06, + "loss": 1.0352, + "step": 4192 + }, + { + "epoch": 0.5041784404497084, + "grad_norm": 1.8369607873842073, + "learning_rate": 2.0700930915599264e-06, + "loss": 0.957, + "step": 4193 + }, + { + "epoch": 0.5042986833403476, + "grad_norm": 2.3619936122824066, + "learning_rate": 2.0693145931527583e-06, + "loss": 1.0216, + "step": 4194 + }, + { + "epoch": 0.5044189262309866, + "grad_norm": 1.6063068364993716, + "learning_rate": 2.068536084230622e-06, + "loss": 1.0181, + "step": 4195 + }, + { + "epoch": 0.5045391691216257, + "grad_norm": 2.4763273975922324, + "learning_rate": 2.067757564911616e-06, + "loss": 1.1276, + "step": 4196 + }, + { + "epoch": 0.5046594120122648, + "grad_norm": 1.9559720460419319, + "learning_rate": 2.0669790353138407e-06, + "loss": 1.1636, + "step": 4197 + }, + { + "epoch": 0.5047796549029039, + "grad_norm": 2.1891551704041956, + "learning_rate": 2.0662004955553995e-06, + "loss": 0.9681, + "step": 4198 + }, + { + "epoch": 0.5048998977935429, + "grad_norm": 2.1478984364557037, + "learning_rate": 2.065421945754395e-06, + "loss": 1.0125, + "step": 4199 + }, + { + "epoch": 0.505020140684182, + "grad_norm": 1.615316294353661, + "learning_rate": 2.0646433860289344e-06, + "loss": 1.0169, + "step": 4200 + }, + { + "epoch": 0.5051403835748212, + "grad_norm": 3.1394283615033673, + "learning_rate": 2.0638648164971233e-06, + "loss": 1.0643, + "step": 4201 + }, + { + "epoch": 0.5052606264654602, + "grad_norm": 5.319623950201519, + "learning_rate": 2.06308623727707e-06, + "loss": 1.1214, + "step": 4202 + }, + { + "epoch": 0.5053808693560993, + "grad_norm": 3.635954255121606, + "learning_rate": 2.0623076484868846e-06, + "loss": 1.0029, + "step": 4203 + }, + { + "epoch": 0.5055011122467384, + "grad_norm": 0.846756065997912, + "learning_rate": 2.061529050244679e-06, + "loss": 0.8451, + "step": 4204 + }, + { + "epoch": 0.5056213551373775, + "grad_norm": 2.100209648021671, + "learning_rate": 2.060750442668565e-06, + "loss": 0.9769, + "step": 4205 + }, + { + "epoch": 0.5057415980280165, + "grad_norm": 2.4271243396181914, + "learning_rate": 2.059971825876657e-06, + "loss": 0.88, + "step": 4206 + }, + { + "epoch": 0.5058618409186557, + "grad_norm": 1.9794459965665556, + "learning_rate": 2.0591931999870713e-06, + "loss": 1.0007, + "step": 4207 + }, + { + "epoch": 0.5059820838092948, + "grad_norm": 0.8259351314491953, + "learning_rate": 2.0584145651179234e-06, + "loss": 0.8136, + "step": 4208 + }, + { + "epoch": 0.5061023266999338, + "grad_norm": 3.0861145741324725, + "learning_rate": 2.0576359213873327e-06, + "loss": 1.034, + "step": 4209 + }, + { + "epoch": 0.506222569590573, + "grad_norm": 2.382359596110172, + "learning_rate": 2.056857268913419e-06, + "loss": 0.9375, + "step": 4210 + }, + { + "epoch": 0.506342812481212, + "grad_norm": 2.2571712229983, + "learning_rate": 2.056078607814303e-06, + "loss": 1.0776, + "step": 4211 + }, + { + "epoch": 0.5064630553718511, + "grad_norm": 1.8881844144966862, + "learning_rate": 2.055299938208106e-06, + "loss": 1.0212, + "step": 4212 + }, + { + "epoch": 0.5065832982624903, + "grad_norm": 2.1564410581115765, + "learning_rate": 2.0545212602129526e-06, + "loss": 1.0989, + "step": 4213 + }, + { + "epoch": 0.5067035411531293, + "grad_norm": 1.9954761363346725, + "learning_rate": 2.0537425739469673e-06, + "loss": 0.9039, + "step": 4214 + }, + { + "epoch": 0.5068237840437684, + "grad_norm": 0.8445253694878307, + "learning_rate": 2.052963879528276e-06, + "loss": 0.8317, + "step": 4215 + }, + { + "epoch": 0.5069440269344075, + "grad_norm": 2.0811493170501723, + "learning_rate": 2.052185177075007e-06, + "loss": 1.0006, + "step": 4216 + }, + { + "epoch": 0.5070642698250466, + "grad_norm": 2.0310104228831785, + "learning_rate": 2.051406466705288e-06, + "loss": 1.0678, + "step": 4217 + }, + { + "epoch": 0.5071845127156857, + "grad_norm": 1.8399862894293766, + "learning_rate": 2.0506277485372486e-06, + "loss": 1.0466, + "step": 4218 + }, + { + "epoch": 0.5073047556063248, + "grad_norm": 2.111090068823811, + "learning_rate": 2.04984902268902e-06, + "loss": 0.9075, + "step": 4219 + }, + { + "epoch": 0.5074249984969639, + "grad_norm": 2.0291598790529792, + "learning_rate": 2.0490702892787345e-06, + "loss": 0.9933, + "step": 4220 + }, + { + "epoch": 0.5075452413876029, + "grad_norm": 2.0053785243615376, + "learning_rate": 2.0482915484245246e-06, + "loss": 0.8573, + "step": 4221 + }, + { + "epoch": 0.5076654842782421, + "grad_norm": 2.3057620889263446, + "learning_rate": 2.047512800244526e-06, + "loss": 1.0734, + "step": 4222 + }, + { + "epoch": 0.5077857271688812, + "grad_norm": 2.465181499863754, + "learning_rate": 2.046734044856873e-06, + "loss": 1.0308, + "step": 4223 + }, + { + "epoch": 0.5079059700595202, + "grad_norm": 1.9025400048760628, + "learning_rate": 2.045955282379702e-06, + "loss": 1.05, + "step": 4224 + }, + { + "epoch": 0.5080262129501594, + "grad_norm": 3.9708041184355674, + "learning_rate": 2.045176512931152e-06, + "loss": 0.9919, + "step": 4225 + }, + { + "epoch": 0.5081464558407984, + "grad_norm": 3.38354202403605, + "learning_rate": 2.0443977366293604e-06, + "loss": 0.9966, + "step": 4226 + }, + { + "epoch": 0.5082666987314375, + "grad_norm": 2.5894438145669096, + "learning_rate": 2.043618953592468e-06, + "loss": 1.0072, + "step": 4227 + }, + { + "epoch": 0.5083869416220766, + "grad_norm": 2.8938603621132373, + "learning_rate": 2.0428401639386144e-06, + "loss": 1.0485, + "step": 4228 + }, + { + "epoch": 0.5085071845127157, + "grad_norm": 0.8277448105345525, + "learning_rate": 2.042061367785943e-06, + "loss": 0.8181, + "step": 4229 + }, + { + "epoch": 0.5086274274033548, + "grad_norm": 2.271027536283222, + "learning_rate": 2.041282565252594e-06, + "loss": 0.9877, + "step": 4230 + }, + { + "epoch": 0.5087476702939938, + "grad_norm": 1.7657697500123264, + "learning_rate": 2.040503756456714e-06, + "loss": 1.0086, + "step": 4231 + }, + { + "epoch": 0.508867913184633, + "grad_norm": 2.302231336528128, + "learning_rate": 2.0397249415164456e-06, + "loss": 1.0243, + "step": 4232 + }, + { + "epoch": 0.508988156075272, + "grad_norm": 2.950452937899082, + "learning_rate": 2.0389461205499354e-06, + "loss": 1.0392, + "step": 4233 + }, + { + "epoch": 0.5091083989659111, + "grad_norm": 1.9531486204626396, + "learning_rate": 2.03816729367533e-06, + "loss": 0.9718, + "step": 4234 + }, + { + "epoch": 0.5092286418565503, + "grad_norm": 6.821387417083515, + "learning_rate": 2.0373884610107765e-06, + "loss": 0.9471, + "step": 4235 + }, + { + "epoch": 0.5093488847471893, + "grad_norm": 2.935716452811578, + "learning_rate": 2.0366096226744225e-06, + "loss": 0.9338, + "step": 4236 + }, + { + "epoch": 0.5094691276378284, + "grad_norm": 1.7941580092378784, + "learning_rate": 2.035830778784418e-06, + "loss": 1.0033, + "step": 4237 + }, + { + "epoch": 0.5095893705284675, + "grad_norm": 1.8825232572499375, + "learning_rate": 2.0350519294589134e-06, + "loss": 1.0343, + "step": 4238 + }, + { + "epoch": 0.5097096134191066, + "grad_norm": 1.8571608366986716, + "learning_rate": 2.0342730748160588e-06, + "loss": 1.0669, + "step": 4239 + }, + { + "epoch": 0.5098298563097456, + "grad_norm": 2.2972014318486633, + "learning_rate": 2.033494214974006e-06, + "loss": 0.9383, + "step": 4240 + }, + { + "epoch": 0.5099500992003848, + "grad_norm": 2.0380076752987013, + "learning_rate": 2.0327153500509067e-06, + "loss": 1.0739, + "step": 4241 + }, + { + "epoch": 0.5100703420910239, + "grad_norm": 2.1644483972761828, + "learning_rate": 2.031936480164916e-06, + "loss": 1.0852, + "step": 4242 + }, + { + "epoch": 0.5101905849816629, + "grad_norm": 1.8055310940511917, + "learning_rate": 2.0311576054341857e-06, + "loss": 1.0326, + "step": 4243 + }, + { + "epoch": 0.5103108278723021, + "grad_norm": 4.24457652789098, + "learning_rate": 2.0303787259768715e-06, + "loss": 0.8657, + "step": 4244 + }, + { + "epoch": 0.5104310707629411, + "grad_norm": 2.352076781550682, + "learning_rate": 2.0295998419111294e-06, + "loss": 0.9284, + "step": 4245 + }, + { + "epoch": 0.5105513136535802, + "grad_norm": 5.575919775934393, + "learning_rate": 2.028820953355115e-06, + "loss": 0.9729, + "step": 4246 + }, + { + "epoch": 0.5106715565442194, + "grad_norm": 1.7847947437017473, + "learning_rate": 2.0280420604269834e-06, + "loss": 1.0233, + "step": 4247 + }, + { + "epoch": 0.5107917994348584, + "grad_norm": 0.7123352814054055, + "learning_rate": 2.027263163244895e-06, + "loss": 0.8268, + "step": 4248 + }, + { + "epoch": 0.5109120423254975, + "grad_norm": 1.6883953686106372, + "learning_rate": 2.026484261927005e-06, + "loss": 0.979, + "step": 4249 + }, + { + "epoch": 0.5110322852161366, + "grad_norm": 4.518741785994329, + "learning_rate": 2.025705356591475e-06, + "loss": 0.9779, + "step": 4250 + }, + { + "epoch": 0.5111525281067757, + "grad_norm": 0.7641761461236423, + "learning_rate": 2.024926447356462e-06, + "loss": 0.8171, + "step": 4251 + }, + { + "epoch": 0.5112727709974147, + "grad_norm": 3.7333419839440896, + "learning_rate": 2.024147534340127e-06, + "loss": 1.0243, + "step": 4252 + }, + { + "epoch": 0.5113930138880539, + "grad_norm": 1.595465839724417, + "learning_rate": 2.02336861766063e-06, + "loss": 1.0346, + "step": 4253 + }, + { + "epoch": 0.511513256778693, + "grad_norm": 1.9945086551496634, + "learning_rate": 2.0225896974361327e-06, + "loss": 1.0286, + "step": 4254 + }, + { + "epoch": 0.511633499669332, + "grad_norm": 0.8652273139062715, + "learning_rate": 2.0218107737847962e-06, + "loss": 0.837, + "step": 4255 + }, + { + "epoch": 0.5117537425599712, + "grad_norm": 2.0267571866792786, + "learning_rate": 2.0210318468247826e-06, + "loss": 0.9866, + "step": 4256 + }, + { + "epoch": 0.5118739854506102, + "grad_norm": 2.1140280935753966, + "learning_rate": 2.020252916674255e-06, + "loss": 1.0585, + "step": 4257 + }, + { + "epoch": 0.5119942283412493, + "grad_norm": 1.9604566372057648, + "learning_rate": 2.019473983451375e-06, + "loss": 1.0504, + "step": 4258 + }, + { + "epoch": 0.5121144712318885, + "grad_norm": 1.7714606202540089, + "learning_rate": 2.0186950472743076e-06, + "loss": 0.9497, + "step": 4259 + }, + { + "epoch": 0.5122347141225275, + "grad_norm": 1.6655131003632753, + "learning_rate": 2.0179161082612162e-06, + "loss": 0.976, + "step": 4260 + }, + { + "epoch": 0.5123549570131666, + "grad_norm": 2.0028980239604723, + "learning_rate": 2.017137166530266e-06, + "loss": 0.9652, + "step": 4261 + }, + { + "epoch": 0.5124751999038056, + "grad_norm": 2.8208004009028, + "learning_rate": 2.0163582221996213e-06, + "loss": 1.0402, + "step": 4262 + }, + { + "epoch": 0.5125954427944448, + "grad_norm": 2.020397952132017, + "learning_rate": 2.015579275387446e-06, + "loss": 0.9164, + "step": 4263 + }, + { + "epoch": 0.5127156856850839, + "grad_norm": 2.0181386488437885, + "learning_rate": 2.0148003262119085e-06, + "loss": 0.9194, + "step": 4264 + }, + { + "epoch": 0.5128359285757229, + "grad_norm": 3.0314639301120363, + "learning_rate": 2.0140213747911728e-06, + "loss": 1.0027, + "step": 4265 + }, + { + "epoch": 0.5129561714663621, + "grad_norm": 2.455610923037397, + "learning_rate": 2.013242421243406e-06, + "loss": 1.0417, + "step": 4266 + }, + { + "epoch": 0.5130764143570011, + "grad_norm": 1.578103849061541, + "learning_rate": 2.012463465686774e-06, + "loss": 1.0251, + "step": 4267 + }, + { + "epoch": 0.5131966572476402, + "grad_norm": 0.7776961558141634, + "learning_rate": 2.0116845082394446e-06, + "loss": 0.7859, + "step": 4268 + }, + { + "epoch": 0.5133169001382794, + "grad_norm": 2.056640161295057, + "learning_rate": 2.0109055490195836e-06, + "loss": 1.0219, + "step": 4269 + }, + { + "epoch": 0.5134371430289184, + "grad_norm": 2.0357807980334233, + "learning_rate": 2.0101265881453605e-06, + "loss": 0.8796, + "step": 4270 + }, + { + "epoch": 0.5135573859195575, + "grad_norm": 1.9796155186357616, + "learning_rate": 2.009347625734941e-06, + "loss": 1.0201, + "step": 4271 + }, + { + "epoch": 0.5136776288101966, + "grad_norm": 4.513887381268111, + "learning_rate": 2.0085686619064954e-06, + "loss": 0.9886, + "step": 4272 + }, + { + "epoch": 0.5137978717008357, + "grad_norm": 2.3512054882064684, + "learning_rate": 2.00778969677819e-06, + "loss": 1.0611, + "step": 4273 + }, + { + "epoch": 0.5139181145914747, + "grad_norm": 1.9998344114419213, + "learning_rate": 2.0070107304681934e-06, + "loss": 0.8732, + "step": 4274 + }, + { + "epoch": 0.5140383574821139, + "grad_norm": 1.9538700971343592, + "learning_rate": 2.006231763094675e-06, + "loss": 1.0168, + "step": 4275 + }, + { + "epoch": 0.514158600372753, + "grad_norm": 1.8732346171143686, + "learning_rate": 2.0054527947758027e-06, + "loss": 1.1067, + "step": 4276 + }, + { + "epoch": 0.514278843263392, + "grad_norm": 0.7366160024795468, + "learning_rate": 2.004673825629746e-06, + "loss": 0.797, + "step": 4277 + }, + { + "epoch": 0.5143990861540312, + "grad_norm": 1.5742717573077825, + "learning_rate": 2.0038948557746744e-06, + "loss": 0.9607, + "step": 4278 + }, + { + "epoch": 0.5145193290446702, + "grad_norm": 1.837839383293709, + "learning_rate": 2.0031158853287558e-06, + "loss": 0.9889, + "step": 4279 + }, + { + "epoch": 0.5146395719353093, + "grad_norm": 2.497259163456563, + "learning_rate": 2.0023369144101593e-06, + "loss": 0.9435, + "step": 4280 + }, + { + "epoch": 0.5147598148259485, + "grad_norm": 1.8616831581981277, + "learning_rate": 2.0015579431370555e-06, + "loss": 1.0014, + "step": 4281 + }, + { + "epoch": 0.5148800577165875, + "grad_norm": 2.7949147094378795, + "learning_rate": 2.000778971627612e-06, + "loss": 0.9363, + "step": 4282 + }, + { + "epoch": 0.5150003006072266, + "grad_norm": 1.8616382704743166, + "learning_rate": 2e-06, + "loss": 1.1423, + "step": 4283 + }, + { + "epoch": 0.5151205434978657, + "grad_norm": 1.7683163276355618, + "learning_rate": 1.9992210283723878e-06, + "loss": 1.0928, + "step": 4284 + }, + { + "epoch": 0.5152407863885048, + "grad_norm": 1.6256521090278793, + "learning_rate": 1.9984420568629448e-06, + "loss": 1.0315, + "step": 4285 + }, + { + "epoch": 0.5153610292791438, + "grad_norm": 2.610400335367867, + "learning_rate": 1.9976630855898405e-06, + "loss": 1.0196, + "step": 4286 + }, + { + "epoch": 0.515481272169783, + "grad_norm": 2.139402165553864, + "learning_rate": 1.9968841146712445e-06, + "loss": 0.9754, + "step": 4287 + }, + { + "epoch": 0.5156015150604221, + "grad_norm": 1.7985780026380525, + "learning_rate": 1.996105144225326e-06, + "loss": 0.9514, + "step": 4288 + }, + { + "epoch": 0.5157217579510611, + "grad_norm": 1.8839487153502088, + "learning_rate": 1.995326174370254e-06, + "loss": 1.0288, + "step": 4289 + }, + { + "epoch": 0.5158420008417003, + "grad_norm": 1.6677213034124476, + "learning_rate": 1.994547205224197e-06, + "loss": 0.9687, + "step": 4290 + }, + { + "epoch": 0.5159622437323393, + "grad_norm": 4.281386101779965, + "learning_rate": 1.993768236905325e-06, + "loss": 0.9144, + "step": 4291 + }, + { + "epoch": 0.5160824866229784, + "grad_norm": 2.7094541333531796, + "learning_rate": 1.992989269531807e-06, + "loss": 0.896, + "step": 4292 + }, + { + "epoch": 0.5162027295136175, + "grad_norm": 2.7576415789754987, + "learning_rate": 1.99221030322181e-06, + "loss": 0.9192, + "step": 4293 + }, + { + "epoch": 0.5163229724042566, + "grad_norm": 1.6906722592783785, + "learning_rate": 1.991431338093505e-06, + "loss": 1.0451, + "step": 4294 + }, + { + "epoch": 0.5164432152948957, + "grad_norm": 1.969463945545423, + "learning_rate": 1.9906523742650587e-06, + "loss": 1.0272, + "step": 4295 + }, + { + "epoch": 0.5165634581855347, + "grad_norm": 2.1521702746702727, + "learning_rate": 1.9898734118546397e-06, + "loss": 1.0102, + "step": 4296 + }, + { + "epoch": 0.5166837010761739, + "grad_norm": 1.6197940626055138, + "learning_rate": 1.989094450980416e-06, + "loss": 1.0385, + "step": 4297 + }, + { + "epoch": 0.516803943966813, + "grad_norm": 2.2067158971675007, + "learning_rate": 1.9883154917605556e-06, + "loss": 1.0028, + "step": 4298 + }, + { + "epoch": 0.516924186857452, + "grad_norm": 2.5074919023316484, + "learning_rate": 1.9875365343132262e-06, + "loss": 1.0691, + "step": 4299 + }, + { + "epoch": 0.5170444297480912, + "grad_norm": 10.820358683904145, + "learning_rate": 1.9867575787565946e-06, + "loss": 1.0886, + "step": 4300 + }, + { + "epoch": 0.5171646726387302, + "grad_norm": 2.111232476621894, + "learning_rate": 1.9859786252088275e-06, + "loss": 1.1003, + "step": 4301 + }, + { + "epoch": 0.5172849155293693, + "grad_norm": 3.395741061651799, + "learning_rate": 1.9851996737880914e-06, + "loss": 0.9042, + "step": 4302 + }, + { + "epoch": 0.5174051584200084, + "grad_norm": 2.2986366626539683, + "learning_rate": 1.9844207246125537e-06, + "loss": 0.9826, + "step": 4303 + }, + { + "epoch": 0.5175254013106475, + "grad_norm": 2.0480084957155205, + "learning_rate": 1.983641777800379e-06, + "loss": 0.9217, + "step": 4304 + }, + { + "epoch": 0.5176456442012866, + "grad_norm": 0.7506626301400569, + "learning_rate": 1.9828628334697343e-06, + "loss": 0.8253, + "step": 4305 + }, + { + "epoch": 0.5177658870919257, + "grad_norm": 0.7746715911179625, + "learning_rate": 1.982083891738784e-06, + "loss": 0.7842, + "step": 4306 + }, + { + "epoch": 0.5178861299825648, + "grad_norm": 1.7496985448320734, + "learning_rate": 1.9813049527256923e-06, + "loss": 1.066, + "step": 4307 + }, + { + "epoch": 0.5180063728732038, + "grad_norm": 3.6719087639737547, + "learning_rate": 1.9805260165486252e-06, + "loss": 1.0518, + "step": 4308 + }, + { + "epoch": 0.518126615763843, + "grad_norm": 2.121269654344221, + "learning_rate": 1.9797470833257457e-06, + "loss": 1.0984, + "step": 4309 + }, + { + "epoch": 0.5182468586544821, + "grad_norm": 1.9935159480076745, + "learning_rate": 1.9789681531752177e-06, + "loss": 1.0076, + "step": 4310 + }, + { + "epoch": 0.5183671015451211, + "grad_norm": 1.5227002829873293, + "learning_rate": 1.978189226215204e-06, + "loss": 0.9599, + "step": 4311 + }, + { + "epoch": 0.5184873444357603, + "grad_norm": 2.1805831545671293, + "learning_rate": 1.9774103025638675e-06, + "loss": 1.0084, + "step": 4312 + }, + { + "epoch": 0.5186075873263993, + "grad_norm": 1.5672571625064389, + "learning_rate": 1.9766313823393696e-06, + "loss": 1.0027, + "step": 4313 + }, + { + "epoch": 0.5187278302170384, + "grad_norm": 2.7540753251788583, + "learning_rate": 1.975852465659873e-06, + "loss": 0.9243, + "step": 4314 + }, + { + "epoch": 0.5188480731076776, + "grad_norm": 2.366758704298725, + "learning_rate": 1.9750735526435377e-06, + "loss": 0.9408, + "step": 4315 + }, + { + "epoch": 0.5189683159983166, + "grad_norm": 3.283602207489703, + "learning_rate": 1.974294643408525e-06, + "loss": 1.0322, + "step": 4316 + }, + { + "epoch": 0.5190885588889557, + "grad_norm": 1.8895055435919013, + "learning_rate": 1.9735157380729947e-06, + "loss": 0.9098, + "step": 4317 + }, + { + "epoch": 0.5192088017795948, + "grad_norm": 2.3704795984703004, + "learning_rate": 1.9727368367551053e-06, + "loss": 1.0822, + "step": 4318 + }, + { + "epoch": 0.5193290446702339, + "grad_norm": 1.9206858111460967, + "learning_rate": 1.9719579395730164e-06, + "loss": 0.921, + "step": 4319 + }, + { + "epoch": 0.5194492875608729, + "grad_norm": 2.2946271211976237, + "learning_rate": 1.9711790466448854e-06, + "loss": 1.172, + "step": 4320 + }, + { + "epoch": 0.5195695304515121, + "grad_norm": 2.8336300040114253, + "learning_rate": 1.9704001580888704e-06, + "loss": 0.9536, + "step": 4321 + }, + { + "epoch": 0.5196897733421512, + "grad_norm": 1.7789492555428292, + "learning_rate": 1.9696212740231283e-06, + "loss": 1.1032, + "step": 4322 + }, + { + "epoch": 0.5198100162327902, + "grad_norm": 2.7714012730339714, + "learning_rate": 1.9688423945658146e-06, + "loss": 1.0552, + "step": 4323 + }, + { + "epoch": 0.5199302591234293, + "grad_norm": 2.6557060133458466, + "learning_rate": 1.9680635198350845e-06, + "loss": 0.9586, + "step": 4324 + }, + { + "epoch": 0.5200505020140684, + "grad_norm": 2.6729083823484237, + "learning_rate": 1.967284649949093e-06, + "loss": 0.9632, + "step": 4325 + }, + { + "epoch": 0.5201707449047075, + "grad_norm": 1.9424739488305567, + "learning_rate": 1.966505785025994e-06, + "loss": 0.9566, + "step": 4326 + }, + { + "epoch": 0.5202909877953465, + "grad_norm": 2.025045926550475, + "learning_rate": 1.965726925183941e-06, + "loss": 1.0017, + "step": 4327 + }, + { + "epoch": 0.5204112306859857, + "grad_norm": 1.972522935045006, + "learning_rate": 1.964948070541087e-06, + "loss": 1.081, + "step": 4328 + }, + { + "epoch": 0.5205314735766248, + "grad_norm": 2.816230503900317, + "learning_rate": 1.9641692212155816e-06, + "loss": 0.9368, + "step": 4329 + }, + { + "epoch": 0.5206517164672638, + "grad_norm": 2.0535018458928165, + "learning_rate": 1.9633903773255777e-06, + "loss": 0.9667, + "step": 4330 + }, + { + "epoch": 0.520771959357903, + "grad_norm": 1.7894024921507752, + "learning_rate": 1.9626115389892237e-06, + "loss": 0.9817, + "step": 4331 + }, + { + "epoch": 0.520892202248542, + "grad_norm": 2.391527192660042, + "learning_rate": 1.96183270632467e-06, + "loss": 1.0939, + "step": 4332 + }, + { + "epoch": 0.5210124451391811, + "grad_norm": 1.6936137746384987, + "learning_rate": 1.9610538794500644e-06, + "loss": 1.0291, + "step": 4333 + }, + { + "epoch": 0.5211326880298203, + "grad_norm": 0.7830036698933779, + "learning_rate": 1.9602750584835542e-06, + "loss": 0.8315, + "step": 4334 + }, + { + "epoch": 0.5212529309204593, + "grad_norm": 2.064740207023043, + "learning_rate": 1.959496243543286e-06, + "loss": 1.0659, + "step": 4335 + }, + { + "epoch": 0.5213731738110984, + "grad_norm": 2.039551774879307, + "learning_rate": 1.9587174347474057e-06, + "loss": 1.029, + "step": 4336 + }, + { + "epoch": 0.5214934167017375, + "grad_norm": 4.293963413404252, + "learning_rate": 1.9579386322140574e-06, + "loss": 1.0506, + "step": 4337 + }, + { + "epoch": 0.5216136595923766, + "grad_norm": 4.322940296277035, + "learning_rate": 1.9571598360613854e-06, + "loss": 1.0464, + "step": 4338 + }, + { + "epoch": 0.5217339024830157, + "grad_norm": 3.751355879756608, + "learning_rate": 1.956381046407532e-06, + "loss": 0.9301, + "step": 4339 + }, + { + "epoch": 0.5218541453736548, + "grad_norm": 1.7743525224090326, + "learning_rate": 1.9556022633706394e-06, + "loss": 1.0995, + "step": 4340 + }, + { + "epoch": 0.5219743882642939, + "grad_norm": 1.822822248647628, + "learning_rate": 1.954823487068848e-06, + "loss": 1.0353, + "step": 4341 + }, + { + "epoch": 0.5220946311549329, + "grad_norm": 1.6429752046800217, + "learning_rate": 1.9540447176202976e-06, + "loss": 1.0472, + "step": 4342 + }, + { + "epoch": 0.5222148740455721, + "grad_norm": 0.8761057338503889, + "learning_rate": 1.9532659551431272e-06, + "loss": 0.8451, + "step": 4343 + }, + { + "epoch": 0.5223351169362112, + "grad_norm": 1.6107815272010642, + "learning_rate": 1.9524871997554744e-06, + "loss": 0.9132, + "step": 4344 + }, + { + "epoch": 0.5224553598268502, + "grad_norm": 2.609142909892665, + "learning_rate": 1.951708451575475e-06, + "loss": 1.043, + "step": 4345 + }, + { + "epoch": 0.5225756027174894, + "grad_norm": 2.3486459077402597, + "learning_rate": 1.9509297107212657e-06, + "loss": 1.0615, + "step": 4346 + }, + { + "epoch": 0.5226958456081284, + "grad_norm": 1.6609764317052478, + "learning_rate": 1.95015097731098e-06, + "loss": 1.0265, + "step": 4347 + }, + { + "epoch": 0.5228160884987675, + "grad_norm": 2.158848620240035, + "learning_rate": 1.949372251462751e-06, + "loss": 1.0521, + "step": 4348 + }, + { + "epoch": 0.5229363313894067, + "grad_norm": 8.080552819102458, + "learning_rate": 1.9485935332947124e-06, + "loss": 1.0668, + "step": 4349 + }, + { + "epoch": 0.5230565742800457, + "grad_norm": 3.786175003190576, + "learning_rate": 1.947814822924993e-06, + "loss": 1.0743, + "step": 4350 + }, + { + "epoch": 0.5231768171706848, + "grad_norm": 1.9550226377069226, + "learning_rate": 1.9470361204717236e-06, + "loss": 1.0681, + "step": 4351 + }, + { + "epoch": 0.5232970600613239, + "grad_norm": 1.650222751724766, + "learning_rate": 1.9462574260530326e-06, + "loss": 1.0434, + "step": 4352 + }, + { + "epoch": 0.523417302951963, + "grad_norm": 2.0360344513112416, + "learning_rate": 1.9454787397870472e-06, + "loss": 1.0456, + "step": 4353 + }, + { + "epoch": 0.523537545842602, + "grad_norm": 2.137781970070006, + "learning_rate": 1.944700061791894e-06, + "loss": 0.9539, + "step": 4354 + }, + { + "epoch": 0.5236577887332411, + "grad_norm": 2.2827101959344396, + "learning_rate": 1.943921392185698e-06, + "loss": 0.8909, + "step": 4355 + }, + { + "epoch": 0.5237780316238803, + "grad_norm": 2.510356242823187, + "learning_rate": 1.9431427310865814e-06, + "loss": 1.0048, + "step": 4356 + }, + { + "epoch": 0.5238982745145193, + "grad_norm": 2.7754569424914566, + "learning_rate": 1.942364078612667e-06, + "loss": 1.0219, + "step": 4357 + }, + { + "epoch": 0.5240185174051584, + "grad_norm": 1.7861079844672703, + "learning_rate": 1.9415854348820765e-06, + "loss": 0.9943, + "step": 4358 + }, + { + "epoch": 0.5241387602957975, + "grad_norm": 3.267896966148086, + "learning_rate": 1.940806800012929e-06, + "loss": 0.9203, + "step": 4359 + }, + { + "epoch": 0.5242590031864366, + "grad_norm": 1.5544349043484225, + "learning_rate": 1.9400281741233432e-06, + "loss": 0.8744, + "step": 4360 + }, + { + "epoch": 0.5243792460770756, + "grad_norm": 0.667868511865977, + "learning_rate": 1.939249557331435e-06, + "loss": 0.7628, + "step": 4361 + }, + { + "epoch": 0.5244994889677148, + "grad_norm": 2.180990069185269, + "learning_rate": 1.938470949755321e-06, + "loss": 0.9675, + "step": 4362 + }, + { + "epoch": 0.5246197318583539, + "grad_norm": 0.8152633771503658, + "learning_rate": 1.937692351513115e-06, + "loss": 0.7944, + "step": 4363 + }, + { + "epoch": 0.5247399747489929, + "grad_norm": 1.718378547063746, + "learning_rate": 1.9369137627229297e-06, + "loss": 1.0443, + "step": 4364 + }, + { + "epoch": 0.5248602176396321, + "grad_norm": 2.4121587956728705, + "learning_rate": 1.936135183502877e-06, + "loss": 1.121, + "step": 4365 + }, + { + "epoch": 0.5249804605302711, + "grad_norm": 2.942254297568873, + "learning_rate": 1.935356613971066e-06, + "loss": 1.0407, + "step": 4366 + }, + { + "epoch": 0.5251007034209102, + "grad_norm": 1.7297270069934545, + "learning_rate": 1.9345780542456047e-06, + "loss": 1.0019, + "step": 4367 + }, + { + "epoch": 0.5252209463115494, + "grad_norm": 2.581178094264145, + "learning_rate": 1.9337995044446007e-06, + "loss": 0.9586, + "step": 4368 + }, + { + "epoch": 0.5253411892021884, + "grad_norm": 2.3637990337028696, + "learning_rate": 1.9330209646861596e-06, + "loss": 1.0398, + "step": 4369 + }, + { + "epoch": 0.5254614320928275, + "grad_norm": 1.643071267289429, + "learning_rate": 1.9322424350883843e-06, + "loss": 1.0128, + "step": 4370 + }, + { + "epoch": 0.5255816749834666, + "grad_norm": 2.7133554893857506, + "learning_rate": 1.931463915769379e-06, + "loss": 1.0228, + "step": 4371 + }, + { + "epoch": 0.5257019178741057, + "grad_norm": 2.3482133201253172, + "learning_rate": 1.930685406847242e-06, + "loss": 0.9773, + "step": 4372 + }, + { + "epoch": 0.5258221607647448, + "grad_norm": 1.6072070880951954, + "learning_rate": 1.9299069084400734e-06, + "loss": 1.0528, + "step": 4373 + }, + { + "epoch": 0.5259424036553839, + "grad_norm": 2.5988736390486444, + "learning_rate": 1.9291284206659717e-06, + "loss": 0.9397, + "step": 4374 + }, + { + "epoch": 0.526062646546023, + "grad_norm": 2.1311047322722856, + "learning_rate": 1.928349943643032e-06, + "loss": 0.95, + "step": 4375 + }, + { + "epoch": 0.526182889436662, + "grad_norm": 1.7236083982645933, + "learning_rate": 1.9275714774893493e-06, + "loss": 1.0525, + "step": 4376 + }, + { + "epoch": 0.5263031323273012, + "grad_norm": 2.4581116937412486, + "learning_rate": 1.9267930223230154e-06, + "loss": 0.9638, + "step": 4377 + }, + { + "epoch": 0.5264233752179402, + "grad_norm": 2.0623104123920206, + "learning_rate": 1.9260145782621224e-06, + "loss": 1.0207, + "step": 4378 + }, + { + "epoch": 0.5265436181085793, + "grad_norm": 2.2511870114136823, + "learning_rate": 1.925236145424758e-06, + "loss": 1.1216, + "step": 4379 + }, + { + "epoch": 0.5266638609992185, + "grad_norm": 0.7020715238972155, + "learning_rate": 1.924457723929012e-06, + "loss": 0.8138, + "step": 4380 + }, + { + "epoch": 0.5267841038898575, + "grad_norm": 1.5632420113141472, + "learning_rate": 1.9236793138929685e-06, + "loss": 1.0663, + "step": 4381 + }, + { + "epoch": 0.5269043467804966, + "grad_norm": 2.0296399117544763, + "learning_rate": 1.9229009154347133e-06, + "loss": 1.0474, + "step": 4382 + }, + { + "epoch": 0.5270245896711357, + "grad_norm": 2.607134742500233, + "learning_rate": 1.922122528672327e-06, + "loss": 1.0415, + "step": 4383 + }, + { + "epoch": 0.5271448325617748, + "grad_norm": 2.410514223789107, + "learning_rate": 1.9213441537238914e-06, + "loss": 1.0187, + "step": 4384 + }, + { + "epoch": 0.5272650754524139, + "grad_norm": 0.8434219429020184, + "learning_rate": 1.920565790707485e-06, + "loss": 0.8108, + "step": 4385 + }, + { + "epoch": 0.527385318343053, + "grad_norm": 2.247848966025088, + "learning_rate": 1.9197874397411853e-06, + "loss": 0.8976, + "step": 4386 + }, + { + "epoch": 0.5275055612336921, + "grad_norm": 4.654257533667133, + "learning_rate": 1.919009100943067e-06, + "loss": 0.9073, + "step": 4387 + }, + { + "epoch": 0.5276258041243311, + "grad_norm": 8.81962827422493, + "learning_rate": 1.9182307744312043e-06, + "loss": 0.895, + "step": 4388 + }, + { + "epoch": 0.5277460470149702, + "grad_norm": 1.7420925234236146, + "learning_rate": 1.9174524603236676e-06, + "loss": 1.003, + "step": 4389 + }, + { + "epoch": 0.5278662899056094, + "grad_norm": 2.87851251733234, + "learning_rate": 1.916674158738527e-06, + "loss": 0.9999, + "step": 4390 + }, + { + "epoch": 0.5279865327962484, + "grad_norm": 1.9799790839814824, + "learning_rate": 1.9158958697938506e-06, + "loss": 0.8371, + "step": 4391 + }, + { + "epoch": 0.5281067756868875, + "grad_norm": 2.823461576189199, + "learning_rate": 1.9151175936077032e-06, + "loss": 1.0977, + "step": 4392 + }, + { + "epoch": 0.5282270185775266, + "grad_norm": 1.6582791835563924, + "learning_rate": 1.9143393302981507e-06, + "loss": 1.0323, + "step": 4393 + }, + { + "epoch": 0.5283472614681657, + "grad_norm": 1.8750610341628715, + "learning_rate": 1.913561079983252e-06, + "loss": 1.0688, + "step": 4394 + }, + { + "epoch": 0.5284675043588047, + "grad_norm": 2.3811804012830673, + "learning_rate": 1.9127828427810693e-06, + "loss": 0.988, + "step": 4395 + }, + { + "epoch": 0.5285877472494439, + "grad_norm": 2.373720426691614, + "learning_rate": 1.9120046188096607e-06, + "loss": 1.0418, + "step": 4396 + }, + { + "epoch": 0.528707990140083, + "grad_norm": 1.9145222637190085, + "learning_rate": 1.9112264081870804e-06, + "loss": 0.9783, + "step": 4397 + }, + { + "epoch": 0.528828233030722, + "grad_norm": 2.2440154702236406, + "learning_rate": 1.9104482110313843e-06, + "loss": 0.9897, + "step": 4398 + }, + { + "epoch": 0.5289484759213612, + "grad_norm": 1.8565901730370318, + "learning_rate": 1.909670027460623e-06, + "loss": 0.9792, + "step": 4399 + }, + { + "epoch": 0.5290687188120002, + "grad_norm": 1.8782108788683094, + "learning_rate": 1.908891857592847e-06, + "loss": 0.953, + "step": 4400 + }, + { + "epoch": 0.5291889617026393, + "grad_norm": 2.5369047885601623, + "learning_rate": 1.9081137015461034e-06, + "loss": 1.138, + "step": 4401 + }, + { + "epoch": 0.5293092045932785, + "grad_norm": 2.454617768966002, + "learning_rate": 1.9073355594384383e-06, + "loss": 1.1399, + "step": 4402 + }, + { + "epoch": 0.5294294474839175, + "grad_norm": 3.00594265768519, + "learning_rate": 1.906557431387895e-06, + "loss": 1.0401, + "step": 4403 + }, + { + "epoch": 0.5295496903745566, + "grad_norm": 1.9437094215008897, + "learning_rate": 1.905779317512516e-06, + "loss": 1.0246, + "step": 4404 + }, + { + "epoch": 0.5296699332651957, + "grad_norm": 2.005352962942296, + "learning_rate": 1.9050012179303385e-06, + "loss": 1.0434, + "step": 4405 + }, + { + "epoch": 0.5297901761558348, + "grad_norm": 2.6086507922959408, + "learning_rate": 1.904223132759401e-06, + "loss": 0.9247, + "step": 4406 + }, + { + "epoch": 0.5299104190464738, + "grad_norm": 2.165285942100851, + "learning_rate": 1.9034450621177383e-06, + "loss": 0.9263, + "step": 4407 + }, + { + "epoch": 0.530030661937113, + "grad_norm": 2.022454689565132, + "learning_rate": 1.9026670061233824e-06, + "loss": 0.9389, + "step": 4408 + }, + { + "epoch": 0.5301509048277521, + "grad_norm": 1.9194884951048719, + "learning_rate": 1.901888964894365e-06, + "loss": 1.0388, + "step": 4409 + }, + { + "epoch": 0.5302711477183911, + "grad_norm": 1.9479744098556933, + "learning_rate": 1.9011109385487134e-06, + "loss": 0.9133, + "step": 4410 + }, + { + "epoch": 0.5303913906090303, + "grad_norm": 3.9966378386929025, + "learning_rate": 1.900332927204454e-06, + "loss": 0.9008, + "step": 4411 + }, + { + "epoch": 0.5305116334996693, + "grad_norm": 1.8404985627301156, + "learning_rate": 1.8995549309796097e-06, + "loss": 1.0035, + "step": 4412 + }, + { + "epoch": 0.5306318763903084, + "grad_norm": 1.8644875514258594, + "learning_rate": 1.8987769499922028e-06, + "loss": 1.0021, + "step": 4413 + }, + { + "epoch": 0.5307521192809476, + "grad_norm": 2.4362649967833963, + "learning_rate": 1.897998984360252e-06, + "loss": 0.9451, + "step": 4414 + }, + { + "epoch": 0.5308723621715866, + "grad_norm": 1.460365224123606, + "learning_rate": 1.897221034201775e-06, + "loss": 1.0188, + "step": 4415 + }, + { + "epoch": 0.5309926050622257, + "grad_norm": 1.5909346603221348, + "learning_rate": 1.8964430996347842e-06, + "loss": 0.9058, + "step": 4416 + }, + { + "epoch": 0.5311128479528648, + "grad_norm": 2.24653698072427, + "learning_rate": 1.8956651807772931e-06, + "loss": 1.0657, + "step": 4417 + }, + { + "epoch": 0.5312330908435039, + "grad_norm": 3.836050867087507, + "learning_rate": 1.8948872777473115e-06, + "loss": 1.0765, + "step": 4418 + }, + { + "epoch": 0.531353333734143, + "grad_norm": 1.8885193756297303, + "learning_rate": 1.8941093906628458e-06, + "loss": 0.8712, + "step": 4419 + }, + { + "epoch": 0.531473576624782, + "grad_norm": 1.868721941989275, + "learning_rate": 1.893331519641902e-06, + "loss": 0.9456, + "step": 4420 + }, + { + "epoch": 0.5315938195154212, + "grad_norm": 2.5638097811703306, + "learning_rate": 1.8925536648024815e-06, + "loss": 0.978, + "step": 4421 + }, + { + "epoch": 0.5317140624060602, + "grad_norm": 2.0165164366113, + "learning_rate": 1.8917758262625849e-06, + "loss": 0.9998, + "step": 4422 + }, + { + "epoch": 0.5318343052966993, + "grad_norm": 1.6894102234437236, + "learning_rate": 1.8909980041402089e-06, + "loss": 1.0485, + "step": 4423 + }, + { + "epoch": 0.5319545481873384, + "grad_norm": 2.853254260574806, + "learning_rate": 1.8902201985533494e-06, + "loss": 0.8974, + "step": 4424 + }, + { + "epoch": 0.5320747910779775, + "grad_norm": 2.2527040551296476, + "learning_rate": 1.8894424096199983e-06, + "loss": 0.9881, + "step": 4425 + }, + { + "epoch": 0.5321950339686166, + "grad_norm": 2.0090361551209095, + "learning_rate": 1.8886646374581463e-06, + "loss": 1.0936, + "step": 4426 + }, + { + "epoch": 0.5323152768592557, + "grad_norm": 1.8868304549294868, + "learning_rate": 1.8878868821857795e-06, + "loss": 0.95, + "step": 4427 + }, + { + "epoch": 0.5324355197498948, + "grad_norm": 2.0737920347325844, + "learning_rate": 1.8871091439208838e-06, + "loss": 0.9881, + "step": 4428 + }, + { + "epoch": 0.5325557626405338, + "grad_norm": 2.359995375159953, + "learning_rate": 1.8863314227814414e-06, + "loss": 1.0115, + "step": 4429 + }, + { + "epoch": 0.532676005531173, + "grad_norm": 5.549930386493318, + "learning_rate": 1.8855537188854313e-06, + "loss": 0.7221, + "step": 4430 + }, + { + "epoch": 0.5327962484218121, + "grad_norm": 1.9362880238591396, + "learning_rate": 1.8847760323508315e-06, + "loss": 1.0218, + "step": 4431 + }, + { + "epoch": 0.5329164913124511, + "grad_norm": 1.8089919506714214, + "learning_rate": 1.883998363295616e-06, + "loss": 0.9967, + "step": 4432 + }, + { + "epoch": 0.5330367342030903, + "grad_norm": 0.8826302196724899, + "learning_rate": 1.8832207118377565e-06, + "loss": 0.8645, + "step": 4433 + }, + { + "epoch": 0.5331569770937293, + "grad_norm": 2.0667006746180725, + "learning_rate": 1.882443078095222e-06, + "loss": 0.9336, + "step": 4434 + }, + { + "epoch": 0.5332772199843684, + "grad_norm": 0.8573807204673438, + "learning_rate": 1.8816654621859794e-06, + "loss": 0.9053, + "step": 4435 + }, + { + "epoch": 0.5333974628750076, + "grad_norm": 2.5156366573087334, + "learning_rate": 1.8808878642279915e-06, + "loss": 0.963, + "step": 4436 + }, + { + "epoch": 0.5335177057656466, + "grad_norm": 2.9786726713208243, + "learning_rate": 1.8801102843392209e-06, + "loss": 0.8887, + "step": 4437 + }, + { + "epoch": 0.5336379486562857, + "grad_norm": 2.537529679298655, + "learning_rate": 1.8793327226376238e-06, + "loss": 1.0907, + "step": 4438 + }, + { + "epoch": 0.5337581915469248, + "grad_norm": 4.02615460249339, + "learning_rate": 1.8785551792411569e-06, + "loss": 1.04, + "step": 4439 + }, + { + "epoch": 0.5338784344375639, + "grad_norm": 2.0101068946540606, + "learning_rate": 1.8777776542677733e-06, + "loss": 1.0651, + "step": 4440 + }, + { + "epoch": 0.5339986773282029, + "grad_norm": 1.9521790311705074, + "learning_rate": 1.8770001478354216e-06, + "loss": 0.9635, + "step": 4441 + }, + { + "epoch": 0.5341189202188421, + "grad_norm": 2.122804068092774, + "learning_rate": 1.8762226600620504e-06, + "loss": 1.0746, + "step": 4442 + }, + { + "epoch": 0.5342391631094812, + "grad_norm": 3.019430496883923, + "learning_rate": 1.8754451910656031e-06, + "loss": 0.8244, + "step": 4443 + }, + { + "epoch": 0.5343594060001202, + "grad_norm": 1.9342125721219638, + "learning_rate": 1.8746677409640212e-06, + "loss": 1.0661, + "step": 4444 + }, + { + "epoch": 0.5344796488907594, + "grad_norm": 1.7650079239621475, + "learning_rate": 1.8738903098752432e-06, + "loss": 1.0828, + "step": 4445 + }, + { + "epoch": 0.5345998917813984, + "grad_norm": 2.062106239416185, + "learning_rate": 1.8731128979172052e-06, + "loss": 0.9685, + "step": 4446 + }, + { + "epoch": 0.5347201346720375, + "grad_norm": 2.5491197674982424, + "learning_rate": 1.8723355052078394e-06, + "loss": 0.9131, + "step": 4447 + }, + { + "epoch": 0.5348403775626767, + "grad_norm": 2.7591659207478902, + "learning_rate": 1.8715581318650765e-06, + "loss": 1.0118, + "step": 4448 + }, + { + "epoch": 0.5349606204533157, + "grad_norm": 3.133073676963758, + "learning_rate": 1.8707807780068422e-06, + "loss": 1.0561, + "step": 4449 + }, + { + "epoch": 0.5350808633439548, + "grad_norm": 2.090938445401609, + "learning_rate": 1.8700034437510611e-06, + "loss": 0.9006, + "step": 4450 + }, + { + "epoch": 0.5352011062345938, + "grad_norm": 2.1292855301118956, + "learning_rate": 1.8692261292156549e-06, + "loss": 1.052, + "step": 4451 + }, + { + "epoch": 0.535321349125233, + "grad_norm": 2.124561040082659, + "learning_rate": 1.8684488345185401e-06, + "loss": 1.0461, + "step": 4452 + }, + { + "epoch": 0.535441592015872, + "grad_norm": 2.647369259351881, + "learning_rate": 1.8676715597776332e-06, + "loss": 1.0255, + "step": 4453 + }, + { + "epoch": 0.5355618349065111, + "grad_norm": 1.9882134986231177, + "learning_rate": 1.8668943051108455e-06, + "loss": 0.9995, + "step": 4454 + }, + { + "epoch": 0.5356820777971503, + "grad_norm": 1.793437289849319, + "learning_rate": 1.8661170706360856e-06, + "loss": 1.0031, + "step": 4455 + }, + { + "epoch": 0.5358023206877893, + "grad_norm": 1.8231919725610453, + "learning_rate": 1.8653398564712594e-06, + "loss": 1.0535, + "step": 4456 + }, + { + "epoch": 0.5359225635784284, + "grad_norm": 1.7027525888349662, + "learning_rate": 1.8645626627342704e-06, + "loss": 1.0598, + "step": 4457 + }, + { + "epoch": 0.5360428064690675, + "grad_norm": 2.4812887923813927, + "learning_rate": 1.8637854895430172e-06, + "loss": 1.0449, + "step": 4458 + }, + { + "epoch": 0.5361630493597066, + "grad_norm": 1.9986747284744233, + "learning_rate": 1.8630083370153978e-06, + "loss": 0.9293, + "step": 4459 + }, + { + "epoch": 0.5362832922503457, + "grad_norm": 0.7522952799964261, + "learning_rate": 1.8622312052693041e-06, + "loss": 0.7911, + "step": 4460 + }, + { + "epoch": 0.5364035351409848, + "grad_norm": 2.8804677302640926, + "learning_rate": 1.8614540944226267e-06, + "loss": 0.9527, + "step": 4461 + }, + { + "epoch": 0.5365237780316239, + "grad_norm": 2.1390372743644757, + "learning_rate": 1.8606770045932537e-06, + "loss": 0.9188, + "step": 4462 + }, + { + "epoch": 0.5366440209222629, + "grad_norm": 2.2832071974875463, + "learning_rate": 1.859899935899068e-06, + "loss": 1.0537, + "step": 4463 + }, + { + "epoch": 0.5367642638129021, + "grad_norm": 1.595494904584408, + "learning_rate": 1.8591228884579506e-06, + "loss": 1.0286, + "step": 4464 + }, + { + "epoch": 0.5368845067035412, + "grad_norm": 1.9734042420835474, + "learning_rate": 1.8583458623877795e-06, + "loss": 1.0565, + "step": 4465 + }, + { + "epoch": 0.5370047495941802, + "grad_norm": 1.8501091899094502, + "learning_rate": 1.8575688578064281e-06, + "loss": 0.9775, + "step": 4466 + }, + { + "epoch": 0.5371249924848194, + "grad_norm": 1.722525732238528, + "learning_rate": 1.8567918748317674e-06, + "loss": 1.004, + "step": 4467 + }, + { + "epoch": 0.5372452353754584, + "grad_norm": 2.113649234005402, + "learning_rate": 1.8560149135816659e-06, + "loss": 1.0623, + "step": 4468 + }, + { + "epoch": 0.5373654782660975, + "grad_norm": 2.4585906933916095, + "learning_rate": 1.8552379741739873e-06, + "loss": 1.0805, + "step": 4469 + }, + { + "epoch": 0.5374857211567367, + "grad_norm": 0.9037613405542358, + "learning_rate": 1.8544610567265935e-06, + "loss": 0.7944, + "step": 4470 + }, + { + "epoch": 0.5376059640473757, + "grad_norm": 1.9214305829054341, + "learning_rate": 1.853684161357341e-06, + "loss": 1.0694, + "step": 4471 + }, + { + "epoch": 0.5377262069380148, + "grad_norm": 1.7369880580722188, + "learning_rate": 1.852907288184085e-06, + "loss": 1.0061, + "step": 4472 + }, + { + "epoch": 0.5378464498286539, + "grad_norm": 2.3094269205949347, + "learning_rate": 1.8521304373246762e-06, + "loss": 0.9382, + "step": 4473 + }, + { + "epoch": 0.537966692719293, + "grad_norm": 2.272288028579709, + "learning_rate": 1.8513536088969626e-06, + "loss": 1.1241, + "step": 4474 + }, + { + "epoch": 0.538086935609932, + "grad_norm": 1.7151158145246828, + "learning_rate": 1.8505768030187884e-06, + "loss": 1.0414, + "step": 4475 + }, + { + "epoch": 0.5382071785005712, + "grad_norm": 2.375833766548744, + "learning_rate": 1.849800019807995e-06, + "loss": 1.0365, + "step": 4476 + }, + { + "epoch": 0.5383274213912103, + "grad_norm": 2.1716347980088297, + "learning_rate": 1.8490232593824186e-06, + "loss": 0.9468, + "step": 4477 + }, + { + "epoch": 0.5384476642818493, + "grad_norm": 1.7255212742553234, + "learning_rate": 1.8482465218598935e-06, + "loss": 1.0874, + "step": 4478 + }, + { + "epoch": 0.5385679071724885, + "grad_norm": 2.0196998274962734, + "learning_rate": 1.8474698073582508e-06, + "loss": 1.0682, + "step": 4479 + }, + { + "epoch": 0.5386881500631275, + "grad_norm": 2.283658467647185, + "learning_rate": 1.8466931159953166e-06, + "loss": 1.1107, + "step": 4480 + }, + { + "epoch": 0.5388083929537666, + "grad_norm": 1.8541509506188174, + "learning_rate": 1.8459164478889158e-06, + "loss": 1.0779, + "step": 4481 + }, + { + "epoch": 0.5389286358444056, + "grad_norm": 1.7356256323302577, + "learning_rate": 1.8451398031568663e-06, + "loss": 0.9935, + "step": 4482 + }, + { + "epoch": 0.5390488787350448, + "grad_norm": 1.6447871863038603, + "learning_rate": 1.844363181916986e-06, + "loss": 0.9872, + "step": 4483 + }, + { + "epoch": 0.5391691216256839, + "grad_norm": 2.0588854704282546, + "learning_rate": 1.8435865842870868e-06, + "loss": 1.065, + "step": 4484 + }, + { + "epoch": 0.5392893645163229, + "grad_norm": 2.026418485796465, + "learning_rate": 1.8428100103849787e-06, + "loss": 0.9566, + "step": 4485 + }, + { + "epoch": 0.5394096074069621, + "grad_norm": 4.79474445954871, + "learning_rate": 1.842033460328467e-06, + "loss": 0.9695, + "step": 4486 + }, + { + "epoch": 0.5395298502976011, + "grad_norm": 2.215567980033926, + "learning_rate": 1.8412569342353541e-06, + "loss": 0.9854, + "step": 4487 + }, + { + "epoch": 0.5396500931882402, + "grad_norm": 2.427384539293089, + "learning_rate": 1.840480432223438e-06, + "loss": 1.0887, + "step": 4488 + }, + { + "epoch": 0.5397703360788794, + "grad_norm": 2.1321465986282444, + "learning_rate": 1.8397039544105131e-06, + "loss": 1.0135, + "step": 4489 + }, + { + "epoch": 0.5398905789695184, + "grad_norm": 1.7058035409646186, + "learning_rate": 1.8389275009143711e-06, + "loss": 0.9338, + "step": 4490 + }, + { + "epoch": 0.5400108218601575, + "grad_norm": 1.8250849299069047, + "learning_rate": 1.8381510718527988e-06, + "loss": 0.9736, + "step": 4491 + }, + { + "epoch": 0.5401310647507966, + "grad_norm": 1.9460948918054786, + "learning_rate": 1.8373746673435812e-06, + "loss": 0.8723, + "step": 4492 + }, + { + "epoch": 0.5402513076414357, + "grad_norm": 1.7004507925691994, + "learning_rate": 1.8365982875044964e-06, + "loss": 1.0264, + "step": 4493 + }, + { + "epoch": 0.5403715505320748, + "grad_norm": 3.355746132306821, + "learning_rate": 1.8358219324533217e-06, + "loss": 0.9931, + "step": 4494 + }, + { + "epoch": 0.5404917934227139, + "grad_norm": 1.6921538711545503, + "learning_rate": 1.8350456023078292e-06, + "loss": 0.9409, + "step": 4495 + }, + { + "epoch": 0.540612036313353, + "grad_norm": 2.405348472648366, + "learning_rate": 1.8342692971857874e-06, + "loss": 1.014, + "step": 4496 + }, + { + "epoch": 0.540732279203992, + "grad_norm": 2.5303454254522117, + "learning_rate": 1.833493017204962e-06, + "loss": 0.9467, + "step": 4497 + }, + { + "epoch": 0.5408525220946312, + "grad_norm": 2.0976389736835115, + "learning_rate": 1.8327167624831134e-06, + "loss": 1.0199, + "step": 4498 + }, + { + "epoch": 0.5409727649852702, + "grad_norm": 1.661409152109221, + "learning_rate": 1.831940533137999e-06, + "loss": 0.9447, + "step": 4499 + }, + { + "epoch": 0.5410930078759093, + "grad_norm": 1.7660515194092623, + "learning_rate": 1.8311643292873718e-06, + "loss": 0.9636, + "step": 4500 + }, + { + "epoch": 0.5412132507665485, + "grad_norm": 1.995363463943591, + "learning_rate": 1.8303881510489818e-06, + "loss": 1.116, + "step": 4501 + }, + { + "epoch": 0.5413334936571875, + "grad_norm": 1.8681470249035228, + "learning_rate": 1.829611998540574e-06, + "loss": 0.9277, + "step": 4502 + }, + { + "epoch": 0.5414537365478266, + "grad_norm": 2.0153976430488685, + "learning_rate": 1.8288358718798914e-06, + "loss": 1.0354, + "step": 4503 + }, + { + "epoch": 0.5415739794384657, + "grad_norm": 1.8053179543815414, + "learning_rate": 1.8280597711846703e-06, + "loss": 0.96, + "step": 4504 + }, + { + "epoch": 0.5416942223291048, + "grad_norm": 3.6239330431582126, + "learning_rate": 1.8272836965726455e-06, + "loss": 1.0707, + "step": 4505 + }, + { + "epoch": 0.5418144652197439, + "grad_norm": 2.1098797229637367, + "learning_rate": 1.8265076481615461e-06, + "loss": 1.0196, + "step": 4506 + }, + { + "epoch": 0.541934708110383, + "grad_norm": 2.392771835270564, + "learning_rate": 1.8257316260690987e-06, + "loss": 1.1108, + "step": 4507 + }, + { + "epoch": 0.5420549510010221, + "grad_norm": 1.6012788847105532, + "learning_rate": 1.8249556304130254e-06, + "loss": 0.9982, + "step": 4508 + }, + { + "epoch": 0.5421751938916611, + "grad_norm": 1.8870064652276077, + "learning_rate": 1.824179661311044e-06, + "loss": 0.918, + "step": 4509 + }, + { + "epoch": 0.5422954367823003, + "grad_norm": 1.9938695650512708, + "learning_rate": 1.823403718880868e-06, + "loss": 1.0389, + "step": 4510 + }, + { + "epoch": 0.5424156796729394, + "grad_norm": 1.6470730454367817, + "learning_rate": 1.822627803240207e-06, + "loss": 0.8984, + "step": 4511 + }, + { + "epoch": 0.5425359225635784, + "grad_norm": 2.4787989004354607, + "learning_rate": 1.8218519145067675e-06, + "loss": 1.0914, + "step": 4512 + }, + { + "epoch": 0.5426561654542175, + "grad_norm": 1.9737870100923014, + "learning_rate": 1.8210760527982508e-06, + "loss": 1.131, + "step": 4513 + }, + { + "epoch": 0.5427764083448566, + "grad_norm": 2.4044035725969533, + "learning_rate": 1.8203002182323552e-06, + "loss": 0.9891, + "step": 4514 + }, + { + "epoch": 0.5428966512354957, + "grad_norm": 1.7822539194176061, + "learning_rate": 1.819524410926773e-06, + "loss": 0.9933, + "step": 4515 + }, + { + "epoch": 0.5430168941261347, + "grad_norm": 2.3944010115733816, + "learning_rate": 1.8187486309991944e-06, + "loss": 1.006, + "step": 4516 + }, + { + "epoch": 0.5431371370167739, + "grad_norm": 1.885369744603906, + "learning_rate": 1.817972878567304e-06, + "loss": 1.0125, + "step": 4517 + }, + { + "epoch": 0.543257379907413, + "grad_norm": 1.828605262728581, + "learning_rate": 1.8171971537487834e-06, + "loss": 1.0022, + "step": 4518 + }, + { + "epoch": 0.543377622798052, + "grad_norm": 2.311244108069808, + "learning_rate": 1.8164214566613093e-06, + "loss": 1.0408, + "step": 4519 + }, + { + "epoch": 0.5434978656886912, + "grad_norm": 2.595422779842993, + "learning_rate": 1.8156457874225547e-06, + "loss": 0.8909, + "step": 4520 + }, + { + "epoch": 0.5436181085793302, + "grad_norm": 1.761539086360007, + "learning_rate": 1.814870146150187e-06, + "loss": 1.0405, + "step": 4521 + }, + { + "epoch": 0.5437383514699693, + "grad_norm": 2.424921986219703, + "learning_rate": 1.814094532961871e-06, + "loss": 1.0208, + "step": 4522 + }, + { + "epoch": 0.5438585943606085, + "grad_norm": 1.905826333915335, + "learning_rate": 1.8133189479752666e-06, + "loss": 1.0742, + "step": 4523 + }, + { + "epoch": 0.5439788372512475, + "grad_norm": 2.079950319760575, + "learning_rate": 1.8125433913080292e-06, + "loss": 1.0545, + "step": 4524 + }, + { + "epoch": 0.5440990801418866, + "grad_norm": 2.291419264414701, + "learning_rate": 1.811767863077811e-06, + "loss": 1.0594, + "step": 4525 + }, + { + "epoch": 0.5442193230325257, + "grad_norm": 1.7305238598730281, + "learning_rate": 1.8109923634022577e-06, + "loss": 1.02, + "step": 4526 + }, + { + "epoch": 0.5443395659231648, + "grad_norm": 2.3488229401019085, + "learning_rate": 1.8102168923990128e-06, + "loss": 1.1074, + "step": 4527 + }, + { + "epoch": 0.5444598088138038, + "grad_norm": 1.8332051824805669, + "learning_rate": 1.809441450185714e-06, + "loss": 1.0358, + "step": 4528 + }, + { + "epoch": 0.544580051704443, + "grad_norm": 2.310626843601076, + "learning_rate": 1.8086660368799958e-06, + "loss": 0.9736, + "step": 4529 + }, + { + "epoch": 0.5447002945950821, + "grad_norm": 1.8788230068618175, + "learning_rate": 1.807890652599488e-06, + "loss": 1.0111, + "step": 4530 + }, + { + "epoch": 0.5448205374857211, + "grad_norm": 2.0296636402351687, + "learning_rate": 1.8071152974618156e-06, + "loss": 1.0625, + "step": 4531 + }, + { + "epoch": 0.5449407803763603, + "grad_norm": 2.031688466798748, + "learning_rate": 1.806339971584599e-06, + "loss": 1.0175, + "step": 4532 + }, + { + "epoch": 0.5450610232669993, + "grad_norm": 1.7126705502418647, + "learning_rate": 1.8055646750854546e-06, + "loss": 1.0895, + "step": 4533 + }, + { + "epoch": 0.5451812661576384, + "grad_norm": 2.364391728058906, + "learning_rate": 1.8047894080819945e-06, + "loss": 1.0582, + "step": 4534 + }, + { + "epoch": 0.5453015090482776, + "grad_norm": 0.732169634299455, + "learning_rate": 1.8040141706918258e-06, + "loss": 0.8696, + "step": 4535 + }, + { + "epoch": 0.5454217519389166, + "grad_norm": 2.5203265686966874, + "learning_rate": 1.8032389630325525e-06, + "loss": 1.0069, + "step": 4536 + }, + { + "epoch": 0.5455419948295557, + "grad_norm": 1.778551634439264, + "learning_rate": 1.8024637852217707e-06, + "loss": 0.9971, + "step": 4537 + }, + { + "epoch": 0.5456622377201948, + "grad_norm": 2.2154454081805803, + "learning_rate": 1.8016886373770766e-06, + "loss": 1.0818, + "step": 4538 + }, + { + "epoch": 0.5457824806108339, + "grad_norm": 1.8465673613213953, + "learning_rate": 1.8009135196160579e-06, + "loss": 1.0255, + "step": 4539 + }, + { + "epoch": 0.545902723501473, + "grad_norm": 1.7163436690717127, + "learning_rate": 1.8001384320563e-06, + "loss": 1.0822, + "step": 4540 + }, + { + "epoch": 0.5460229663921121, + "grad_norm": 0.7777767796358637, + "learning_rate": 1.7993633748153833e-06, + "loss": 0.8154, + "step": 4541 + }, + { + "epoch": 0.5461432092827512, + "grad_norm": 2.006265600981617, + "learning_rate": 1.7985883480108834e-06, + "loss": 0.9665, + "step": 4542 + }, + { + "epoch": 0.5462634521733902, + "grad_norm": 1.6368978397834035, + "learning_rate": 1.797813351760371e-06, + "loss": 0.9598, + "step": 4543 + }, + { + "epoch": 0.5463836950640293, + "grad_norm": 1.7445180725884637, + "learning_rate": 1.7970383861814116e-06, + "loss": 1.0167, + "step": 4544 + }, + { + "epoch": 0.5465039379546685, + "grad_norm": 1.9543283646414287, + "learning_rate": 1.7962634513915684e-06, + "loss": 0.9788, + "step": 4545 + }, + { + "epoch": 0.5466241808453075, + "grad_norm": 1.8840724164404736, + "learning_rate": 1.7954885475083969e-06, + "loss": 1.0315, + "step": 4546 + }, + { + "epoch": 0.5467444237359466, + "grad_norm": 2.4908378557981155, + "learning_rate": 1.7947136746494513e-06, + "loss": 0.9629, + "step": 4547 + }, + { + "epoch": 0.5468646666265857, + "grad_norm": 2.26497056651412, + "learning_rate": 1.793938832932277e-06, + "loss": 1.1169, + "step": 4548 + }, + { + "epoch": 0.5469849095172248, + "grad_norm": 2.1137667678176957, + "learning_rate": 1.7931640224744185e-06, + "loss": 0.9406, + "step": 4549 + }, + { + "epoch": 0.5471051524078638, + "grad_norm": 2.2347194166095394, + "learning_rate": 1.7923892433934127e-06, + "loss": 0.972, + "step": 4550 + }, + { + "epoch": 0.547225395298503, + "grad_norm": 1.8412641804674144, + "learning_rate": 1.7916144958067939e-06, + "loss": 1.0286, + "step": 4551 + }, + { + "epoch": 0.5473456381891421, + "grad_norm": 2.6973506574327395, + "learning_rate": 1.7908397798320905e-06, + "loss": 1.0226, + "step": 4552 + }, + { + "epoch": 0.5474658810797811, + "grad_norm": 2.2330738894327933, + "learning_rate": 1.7900650955868265e-06, + "loss": 0.9865, + "step": 4553 + }, + { + "epoch": 0.5475861239704203, + "grad_norm": 1.3667574832949334, + "learning_rate": 1.7892904431885202e-06, + "loss": 0.9995, + "step": 4554 + }, + { + "epoch": 0.5477063668610593, + "grad_norm": 1.9493965878136725, + "learning_rate": 1.788515822754686e-06, + "loss": 0.9908, + "step": 4555 + }, + { + "epoch": 0.5478266097516984, + "grad_norm": 2.014580271053518, + "learning_rate": 1.7877412344028335e-06, + "loss": 1.0199, + "step": 4556 + }, + { + "epoch": 0.5479468526423376, + "grad_norm": 2.663906327757263, + "learning_rate": 1.7869666782504668e-06, + "loss": 1.0098, + "step": 4557 + }, + { + "epoch": 0.5480670955329766, + "grad_norm": 1.8370598791431636, + "learning_rate": 1.7861921544150867e-06, + "loss": 0.9232, + "step": 4558 + }, + { + "epoch": 0.5481873384236157, + "grad_norm": 1.9478264310067182, + "learning_rate": 1.7854176630141856e-06, + "loss": 0.9975, + "step": 4559 + }, + { + "epoch": 0.5483075813142548, + "grad_norm": 2.4789211461496965, + "learning_rate": 1.784643204165255e-06, + "loss": 1.083, + "step": 4560 + }, + { + "epoch": 0.5484278242048939, + "grad_norm": 2.158432339885371, + "learning_rate": 1.7838687779857783e-06, + "loss": 1.0103, + "step": 4561 + }, + { + "epoch": 0.5485480670955329, + "grad_norm": 2.2753621242850244, + "learning_rate": 1.7830943845932366e-06, + "loss": 0.8791, + "step": 4562 + }, + { + "epoch": 0.5486683099861721, + "grad_norm": 1.5629855355235958, + "learning_rate": 1.7823200241051044e-06, + "loss": 0.988, + "step": 4563 + }, + { + "epoch": 0.5487885528768112, + "grad_norm": 3.2022726273950464, + "learning_rate": 1.7815456966388513e-06, + "loss": 1.0413, + "step": 4564 + }, + { + "epoch": 0.5489087957674502, + "grad_norm": 2.0874310693381664, + "learning_rate": 1.780771402311943e-06, + "loss": 1.0453, + "step": 4565 + }, + { + "epoch": 0.5490290386580894, + "grad_norm": 1.7364037822292906, + "learning_rate": 1.7799971412418374e-06, + "loss": 1.0244, + "step": 4566 + }, + { + "epoch": 0.5491492815487284, + "grad_norm": 2.214980563558547, + "learning_rate": 1.7792229135459918e-06, + "loss": 0.9795, + "step": 4567 + }, + { + "epoch": 0.5492695244393675, + "grad_norm": 0.7435410454482838, + "learning_rate": 1.7784487193418538e-06, + "loss": 0.854, + "step": 4568 + }, + { + "epoch": 0.5493897673300067, + "grad_norm": 3.454917615026954, + "learning_rate": 1.7776745587468698e-06, + "loss": 0.8468, + "step": 4569 + }, + { + "epoch": 0.5495100102206457, + "grad_norm": 2.4651805327194447, + "learning_rate": 1.7769004318784776e-06, + "loss": 1.0559, + "step": 4570 + }, + { + "epoch": 0.5496302531112848, + "grad_norm": 1.9054809019382397, + "learning_rate": 1.776126338854113e-06, + "loss": 1.0466, + "step": 4571 + }, + { + "epoch": 0.5497504960019239, + "grad_norm": 2.92959895699533, + "learning_rate": 1.7753522797912044e-06, + "loss": 1.0837, + "step": 4572 + }, + { + "epoch": 0.549870738892563, + "grad_norm": 2.2487333759162005, + "learning_rate": 1.7745782548071765e-06, + "loss": 0.9372, + "step": 4573 + }, + { + "epoch": 0.549990981783202, + "grad_norm": 1.922544657333644, + "learning_rate": 1.7738042640194482e-06, + "loss": 0.9782, + "step": 4574 + }, + { + "epoch": 0.5501112246738411, + "grad_norm": 2.3934273843517877, + "learning_rate": 1.7730303075454335e-06, + "loss": 0.9421, + "step": 4575 + }, + { + "epoch": 0.5502314675644803, + "grad_norm": 2.0044887952674575, + "learning_rate": 1.7722563855025402e-06, + "loss": 1.0848, + "step": 4576 + }, + { + "epoch": 0.5503517104551193, + "grad_norm": 2.1129947827803277, + "learning_rate": 1.7714824980081721e-06, + "loss": 0.9442, + "step": 4577 + }, + { + "epoch": 0.5504719533457584, + "grad_norm": 3.1744016909617874, + "learning_rate": 1.7707086451797276e-06, + "loss": 0.973, + "step": 4578 + }, + { + "epoch": 0.5505921962363975, + "grad_norm": 0.7095156693989786, + "learning_rate": 1.7699348271345993e-06, + "loss": 0.7607, + "step": 4579 + }, + { + "epoch": 0.5507124391270366, + "grad_norm": 0.717927794175599, + "learning_rate": 1.7691610439901753e-06, + "loss": 0.7819, + "step": 4580 + }, + { + "epoch": 0.5508326820176757, + "grad_norm": 2.080480173146212, + "learning_rate": 1.7683872958638367e-06, + "loss": 0.9903, + "step": 4581 + }, + { + "epoch": 0.5509529249083148, + "grad_norm": 2.0204642702760567, + "learning_rate": 1.7676135828729614e-06, + "loss": 1.0819, + "step": 4582 + }, + { + "epoch": 0.5510731677989539, + "grad_norm": 1.829394950901269, + "learning_rate": 1.7668399051349205e-06, + "loss": 1.0649, + "step": 4583 + }, + { + "epoch": 0.5511934106895929, + "grad_norm": 2.1541079510102814, + "learning_rate": 1.766066262767081e-06, + "loss": 1.0704, + "step": 4584 + }, + { + "epoch": 0.5513136535802321, + "grad_norm": 2.0678220344497746, + "learning_rate": 1.765292655886803e-06, + "loss": 1.0118, + "step": 4585 + }, + { + "epoch": 0.5514338964708712, + "grad_norm": 1.8206848287655448, + "learning_rate": 1.764519084611443e-06, + "loss": 0.9439, + "step": 4586 + }, + { + "epoch": 0.5515541393615102, + "grad_norm": 1.9108325096602174, + "learning_rate": 1.7637455490583505e-06, + "loss": 1.0141, + "step": 4587 + }, + { + "epoch": 0.5516743822521494, + "grad_norm": 2.005006008757532, + "learning_rate": 1.7629720493448701e-06, + "loss": 1.011, + "step": 4588 + }, + { + "epoch": 0.5517946251427884, + "grad_norm": 2.3260304290276803, + "learning_rate": 1.7621985855883418e-06, + "loss": 1.0901, + "step": 4589 + }, + { + "epoch": 0.5519148680334275, + "grad_norm": 1.8557422958310716, + "learning_rate": 1.7614251579060983e-06, + "loss": 0.9609, + "step": 4590 + }, + { + "epoch": 0.5520351109240667, + "grad_norm": 1.7840064955965174, + "learning_rate": 1.76065176641547e-06, + "loss": 1.0835, + "step": 4591 + }, + { + "epoch": 0.5521553538147057, + "grad_norm": 1.7675162794558583, + "learning_rate": 1.759878411233777e-06, + "loss": 1.0138, + "step": 4592 + }, + { + "epoch": 0.5522755967053448, + "grad_norm": 2.147708505582443, + "learning_rate": 1.7591050924783388e-06, + "loss": 0.9949, + "step": 4593 + }, + { + "epoch": 0.5523958395959839, + "grad_norm": 0.851443991157986, + "learning_rate": 1.7583318102664661e-06, + "loss": 0.8173, + "step": 4594 + }, + { + "epoch": 0.552516082486623, + "grad_norm": 4.573255429359699, + "learning_rate": 1.757558564715466e-06, + "loss": 1.028, + "step": 4595 + }, + { + "epoch": 0.552636325377262, + "grad_norm": 2.9720665325454867, + "learning_rate": 1.7567853559426386e-06, + "loss": 0.9737, + "step": 4596 + }, + { + "epoch": 0.5527565682679012, + "grad_norm": 4.433409834084596, + "learning_rate": 1.7560121840652797e-06, + "loss": 0.9911, + "step": 4597 + }, + { + "epoch": 0.5528768111585403, + "grad_norm": 1.9103523262973245, + "learning_rate": 1.7552390492006782e-06, + "loss": 0.9269, + "step": 4598 + }, + { + "epoch": 0.5529970540491793, + "grad_norm": 1.6621770788942922, + "learning_rate": 1.7544659514661184e-06, + "loss": 0.8937, + "step": 4599 + }, + { + "epoch": 0.5531172969398185, + "grad_norm": 1.9704221254809435, + "learning_rate": 1.7536928909788786e-06, + "loss": 1.0327, + "step": 4600 + }, + { + "epoch": 0.5532375398304575, + "grad_norm": 0.8950866907107276, + "learning_rate": 1.752919867856231e-06, + "loss": 0.8572, + "step": 4601 + }, + { + "epoch": 0.5533577827210966, + "grad_norm": 2.1649490909008984, + "learning_rate": 1.7521468822154436e-06, + "loss": 1.0274, + "step": 4602 + }, + { + "epoch": 0.5534780256117358, + "grad_norm": 1.8812934794423322, + "learning_rate": 1.751373934173777e-06, + "loss": 0.9842, + "step": 4603 + }, + { + "epoch": 0.5535982685023748, + "grad_norm": 1.6884987312834008, + "learning_rate": 1.750601023848487e-06, + "loss": 0.9684, + "step": 4604 + }, + { + "epoch": 0.5537185113930139, + "grad_norm": 1.9272231455080098, + "learning_rate": 1.749828151356823e-06, + "loss": 0.9689, + "step": 4605 + }, + { + "epoch": 0.553838754283653, + "grad_norm": 2.184409983604485, + "learning_rate": 1.7490553168160297e-06, + "loss": 0.9926, + "step": 4606 + }, + { + "epoch": 0.5539589971742921, + "grad_norm": 2.2009864329604967, + "learning_rate": 1.748282520343345e-06, + "loss": 0.9987, + "step": 4607 + }, + { + "epoch": 0.5540792400649311, + "grad_norm": 2.0472500290082825, + "learning_rate": 1.7475097620560023e-06, + "loss": 1.0236, + "step": 4608 + }, + { + "epoch": 0.5541994829555702, + "grad_norm": 2.3716668280116457, + "learning_rate": 1.746737042071228e-06, + "loss": 0.9469, + "step": 4609 + }, + { + "epoch": 0.5543197258462094, + "grad_norm": 1.9307930439694325, + "learning_rate": 1.7459643605062424e-06, + "loss": 1.0248, + "step": 4610 + }, + { + "epoch": 0.5544399687368484, + "grad_norm": 1.5986771061842495, + "learning_rate": 1.745191717478262e-06, + "loss": 1.0425, + "step": 4611 + }, + { + "epoch": 0.5545602116274875, + "grad_norm": 2.065023323647816, + "learning_rate": 1.7444191131044948e-06, + "loss": 1.0354, + "step": 4612 + }, + { + "epoch": 0.5546804545181266, + "grad_norm": 1.8209404905840314, + "learning_rate": 1.7436465475021456e-06, + "loss": 0.9686, + "step": 4613 + }, + { + "epoch": 0.5548006974087657, + "grad_norm": 2.0118849009815056, + "learning_rate": 1.7428740207884111e-06, + "loss": 0.9511, + "step": 4614 + }, + { + "epoch": 0.5549209402994048, + "grad_norm": 2.1359739243276623, + "learning_rate": 1.7421015330804833e-06, + "loss": 0.8468, + "step": 4615 + }, + { + "epoch": 0.5550411831900439, + "grad_norm": 1.9725662664153467, + "learning_rate": 1.7413290844955475e-06, + "loss": 0.9549, + "step": 4616 + }, + { + "epoch": 0.555161426080683, + "grad_norm": 1.8713960343676321, + "learning_rate": 1.7405566751507843e-06, + "loss": 1.0213, + "step": 4617 + }, + { + "epoch": 0.555281668971322, + "grad_norm": 1.7802052027044415, + "learning_rate": 1.7397843051633668e-06, + "loss": 0.911, + "step": 4618 + }, + { + "epoch": 0.5554019118619612, + "grad_norm": 1.6860844361598846, + "learning_rate": 1.739011974650464e-06, + "loss": 0.9502, + "step": 4619 + }, + { + "epoch": 0.5555221547526003, + "grad_norm": 4.513532212992362, + "learning_rate": 1.7382396837292365e-06, + "loss": 1.0027, + "step": 4620 + }, + { + "epoch": 0.5556423976432393, + "grad_norm": 1.7228260605346024, + "learning_rate": 1.737467432516841e-06, + "loss": 0.972, + "step": 4621 + }, + { + "epoch": 0.5557626405338785, + "grad_norm": 2.6383312126853196, + "learning_rate": 1.7366952211304274e-06, + "loss": 0.9793, + "step": 4622 + }, + { + "epoch": 0.5558828834245175, + "grad_norm": 2.4583628916714506, + "learning_rate": 1.735923049687139e-06, + "loss": 1.0773, + "step": 4623 + }, + { + "epoch": 0.5560031263151566, + "grad_norm": 1.478243642807853, + "learning_rate": 1.7351509183041144e-06, + "loss": 0.9742, + "step": 4624 + }, + { + "epoch": 0.5561233692057957, + "grad_norm": 1.898170232601279, + "learning_rate": 1.7343788270984852e-06, + "loss": 0.9545, + "step": 4625 + }, + { + "epoch": 0.5562436120964348, + "grad_norm": 2.398017970866489, + "learning_rate": 1.7336067761873764e-06, + "loss": 0.9846, + "step": 4626 + }, + { + "epoch": 0.5563638549870739, + "grad_norm": 1.9137250330297395, + "learning_rate": 1.7328347656879076e-06, + "loss": 0.9987, + "step": 4627 + }, + { + "epoch": 0.556484097877713, + "grad_norm": 2.851163987832032, + "learning_rate": 1.7320627957171927e-06, + "loss": 0.9194, + "step": 4628 + }, + { + "epoch": 0.5566043407683521, + "grad_norm": 2.060999353274522, + "learning_rate": 1.7312908663923382e-06, + "loss": 1.0515, + "step": 4629 + }, + { + "epoch": 0.5567245836589911, + "grad_norm": 2.0102287033313564, + "learning_rate": 1.7305189778304463e-06, + "loss": 0.9106, + "step": 4630 + }, + { + "epoch": 0.5568448265496303, + "grad_norm": 1.890404176032466, + "learning_rate": 1.729747130148611e-06, + "loss": 1.0374, + "step": 4631 + }, + { + "epoch": 0.5569650694402694, + "grad_norm": 2.404402977642068, + "learning_rate": 1.7289753234639208e-06, + "loss": 1.0068, + "step": 4632 + }, + { + "epoch": 0.5570853123309084, + "grad_norm": 1.8625832455868174, + "learning_rate": 1.7282035578934592e-06, + "loss": 1.0048, + "step": 4633 + }, + { + "epoch": 0.5572055552215476, + "grad_norm": 1.9268875503749983, + "learning_rate": 1.727431833554301e-06, + "loss": 1.0256, + "step": 4634 + }, + { + "epoch": 0.5573257981121866, + "grad_norm": 2.031731416433518, + "learning_rate": 1.7266601505635175e-06, + "loss": 1.0077, + "step": 4635 + }, + { + "epoch": 0.5574460410028257, + "grad_norm": 2.2337867656087913, + "learning_rate": 1.7258885090381717e-06, + "loss": 0.9912, + "step": 4636 + }, + { + "epoch": 0.5575662838934649, + "grad_norm": 2.550684134020555, + "learning_rate": 1.7251169090953213e-06, + "loss": 1.0231, + "step": 4637 + }, + { + "epoch": 0.5576865267841039, + "grad_norm": 2.6891247473366278, + "learning_rate": 1.7243453508520168e-06, + "loss": 1.002, + "step": 4638 + }, + { + "epoch": 0.557806769674743, + "grad_norm": 2.06539805863785, + "learning_rate": 1.7235738344253038e-06, + "loss": 1.0843, + "step": 4639 + }, + { + "epoch": 0.557927012565382, + "grad_norm": 1.9429795090423867, + "learning_rate": 1.72280235993222e-06, + "loss": 1.0611, + "step": 4640 + }, + { + "epoch": 0.5580472554560212, + "grad_norm": 2.5091890259042873, + "learning_rate": 1.722030927489798e-06, + "loss": 0.9286, + "step": 4641 + }, + { + "epoch": 0.5581674983466602, + "grad_norm": 1.888582308360499, + "learning_rate": 1.7212595372150634e-06, + "loss": 0.9805, + "step": 4642 + }, + { + "epoch": 0.5582877412372993, + "grad_norm": 2.480357538568688, + "learning_rate": 1.720488189225035e-06, + "loss": 0.9647, + "step": 4643 + }, + { + "epoch": 0.5584079841279385, + "grad_norm": 3.8932289081007787, + "learning_rate": 1.7197168836367265e-06, + "loss": 1.037, + "step": 4644 + }, + { + "epoch": 0.5585282270185775, + "grad_norm": 1.87338390639564, + "learning_rate": 1.7189456205671433e-06, + "loss": 1.0566, + "step": 4645 + }, + { + "epoch": 0.5586484699092166, + "grad_norm": 2.1173258943827427, + "learning_rate": 1.7181744001332866e-06, + "loss": 1.0573, + "step": 4646 + }, + { + "epoch": 0.5587687127998557, + "grad_norm": 1.907710282120378, + "learning_rate": 1.7174032224521493e-06, + "loss": 0.8678, + "step": 4647 + }, + { + "epoch": 0.5588889556904948, + "grad_norm": 1.632644425644224, + "learning_rate": 1.7166320876407184e-06, + "loss": 0.9355, + "step": 4648 + }, + { + "epoch": 0.5590091985811338, + "grad_norm": 2.320450184651298, + "learning_rate": 1.7158609958159742e-06, + "loss": 0.9162, + "step": 4649 + }, + { + "epoch": 0.559129441471773, + "grad_norm": 2.4097233264442965, + "learning_rate": 1.7150899470948911e-06, + "loss": 1.0232, + "step": 4650 + }, + { + "epoch": 0.5592496843624121, + "grad_norm": 0.8118486728173157, + "learning_rate": 1.7143189415944365e-06, + "loss": 0.8043, + "step": 4651 + }, + { + "epoch": 0.5593699272530511, + "grad_norm": 1.765490433957899, + "learning_rate": 1.7135479794315714e-06, + "loss": 0.9996, + "step": 4652 + }, + { + "epoch": 0.5594901701436903, + "grad_norm": 2.017712126101223, + "learning_rate": 1.7127770607232502e-06, + "loss": 1.027, + "step": 4653 + }, + { + "epoch": 0.5596104130343293, + "grad_norm": 1.872803609601909, + "learning_rate": 1.7120061855864204e-06, + "loss": 1.0334, + "step": 4654 + }, + { + "epoch": 0.5597306559249684, + "grad_norm": 2.0011346697769374, + "learning_rate": 1.7112353541380233e-06, + "loss": 0.9474, + "step": 4655 + }, + { + "epoch": 0.5598508988156076, + "grad_norm": 1.7699214008358446, + "learning_rate": 1.7104645664949931e-06, + "loss": 0.9615, + "step": 4656 + }, + { + "epoch": 0.5599711417062466, + "grad_norm": 1.9488249784265663, + "learning_rate": 1.7096938227742584e-06, + "loss": 0.9525, + "step": 4657 + }, + { + "epoch": 0.5600913845968857, + "grad_norm": 1.9609137909812355, + "learning_rate": 1.70892312309274e-06, + "loss": 1.0775, + "step": 4658 + }, + { + "epoch": 0.5602116274875248, + "grad_norm": 2.7440327144596695, + "learning_rate": 1.7081524675673523e-06, + "loss": 0.9216, + "step": 4659 + }, + { + "epoch": 0.5603318703781639, + "grad_norm": 0.7847449520686391, + "learning_rate": 1.7073818563150026e-06, + "loss": 0.8336, + "step": 4660 + }, + { + "epoch": 0.560452113268803, + "grad_norm": 2.268978189389616, + "learning_rate": 1.7066112894525935e-06, + "loss": 1.103, + "step": 4661 + }, + { + "epoch": 0.5605723561594421, + "grad_norm": 1.9997025507035497, + "learning_rate": 1.7058407670970177e-06, + "loss": 0.9607, + "step": 4662 + }, + { + "epoch": 0.5606925990500812, + "grad_norm": 2.2603830666103044, + "learning_rate": 1.7050702893651643e-06, + "loss": 0.8488, + "step": 4663 + }, + { + "epoch": 0.5608128419407202, + "grad_norm": 2.2499983045783543, + "learning_rate": 1.7042998563739134e-06, + "loss": 0.993, + "step": 4664 + }, + { + "epoch": 0.5609330848313594, + "grad_norm": 1.9367907056409774, + "learning_rate": 1.703529468240139e-06, + "loss": 0.9535, + "step": 4665 + }, + { + "epoch": 0.5610533277219985, + "grad_norm": 2.3477594286290913, + "learning_rate": 1.7027591250807088e-06, + "loss": 0.9768, + "step": 4666 + }, + { + "epoch": 0.5611735706126375, + "grad_norm": 2.3546637176345437, + "learning_rate": 1.7019888270124825e-06, + "loss": 1.0781, + "step": 4667 + }, + { + "epoch": 0.5612938135032767, + "grad_norm": 2.3070309895722954, + "learning_rate": 1.7012185741523147e-06, + "loss": 1.0549, + "step": 4668 + }, + { + "epoch": 0.5614140563939157, + "grad_norm": 1.9906978048907193, + "learning_rate": 1.7004483666170514e-06, + "loss": 0.8647, + "step": 4669 + }, + { + "epoch": 0.5615342992845548, + "grad_norm": 1.973569209278678, + "learning_rate": 1.699678204523533e-06, + "loss": 1.0381, + "step": 4670 + }, + { + "epoch": 0.5616545421751938, + "grad_norm": 4.356420684929548, + "learning_rate": 1.6989080879885918e-06, + "loss": 0.9253, + "step": 4671 + }, + { + "epoch": 0.561774785065833, + "grad_norm": 0.8929430259543749, + "learning_rate": 1.6981380171290544e-06, + "loss": 0.8473, + "step": 4672 + }, + { + "epoch": 0.5618950279564721, + "grad_norm": 2.159817605265013, + "learning_rate": 1.6973679920617396e-06, + "loss": 0.9837, + "step": 4673 + }, + { + "epoch": 0.5620152708471111, + "grad_norm": 2.427369216868839, + "learning_rate": 1.6965980129034603e-06, + "loss": 1.0911, + "step": 4674 + }, + { + "epoch": 0.5621355137377503, + "grad_norm": 1.7889652808814767, + "learning_rate": 1.6958280797710209e-06, + "loss": 0.9988, + "step": 4675 + }, + { + "epoch": 0.5622557566283893, + "grad_norm": 0.7208528854129267, + "learning_rate": 1.6950581927812198e-06, + "loss": 0.7851, + "step": 4676 + }, + { + "epoch": 0.5623759995190284, + "grad_norm": 2.2627943574567673, + "learning_rate": 1.6942883520508486e-06, + "loss": 1.0254, + "step": 4677 + }, + { + "epoch": 0.5624962424096676, + "grad_norm": 2.2995649921533436, + "learning_rate": 1.693518557696691e-06, + "loss": 1.0082, + "step": 4678 + }, + { + "epoch": 0.5626164853003066, + "grad_norm": 2.265241498388308, + "learning_rate": 1.6927488098355252e-06, + "loss": 1.1294, + "step": 4679 + }, + { + "epoch": 0.5627367281909457, + "grad_norm": 0.9291158890356379, + "learning_rate": 1.6919791085841201e-06, + "loss": 0.8694, + "step": 4680 + }, + { + "epoch": 0.5628569710815848, + "grad_norm": 2.301843472466062, + "learning_rate": 1.6912094540592396e-06, + "loss": 1.0258, + "step": 4681 + }, + { + "epoch": 0.5629772139722239, + "grad_norm": 3.5565575009163313, + "learning_rate": 1.6904398463776393e-06, + "loss": 1.0465, + "step": 4682 + }, + { + "epoch": 0.5630974568628629, + "grad_norm": 1.782245023448908, + "learning_rate": 1.6896702856560683e-06, + "loss": 0.9649, + "step": 4683 + }, + { + "epoch": 0.5632176997535021, + "grad_norm": 2.669407290560575, + "learning_rate": 1.6889007720112677e-06, + "loss": 0.9294, + "step": 4684 + }, + { + "epoch": 0.5633379426441412, + "grad_norm": 1.656905566336119, + "learning_rate": 1.6881313055599734e-06, + "loss": 1.0069, + "step": 4685 + }, + { + "epoch": 0.5634581855347802, + "grad_norm": 2.466444846652686, + "learning_rate": 1.6873618864189117e-06, + "loss": 1.057, + "step": 4686 + }, + { + "epoch": 0.5635784284254194, + "grad_norm": 4.0728407492558345, + "learning_rate": 1.686592514704803e-06, + "loss": 1.0154, + "step": 4687 + }, + { + "epoch": 0.5636986713160584, + "grad_norm": 4.760372281302356, + "learning_rate": 1.685823190534361e-06, + "loss": 0.9395, + "step": 4688 + }, + { + "epoch": 0.5638189142066975, + "grad_norm": 2.1080226377759166, + "learning_rate": 1.6850539140242907e-06, + "loss": 1.0701, + "step": 4689 + }, + { + "epoch": 0.5639391570973367, + "grad_norm": 1.9995458802599164, + "learning_rate": 1.684284685291292e-06, + "loss": 1.0566, + "step": 4690 + }, + { + "epoch": 0.5640593999879757, + "grad_norm": 2.268368027323885, + "learning_rate": 1.683515504452055e-06, + "loss": 1.0511, + "step": 4691 + }, + { + "epoch": 0.5641796428786148, + "grad_norm": 1.806968667430535, + "learning_rate": 1.6827463716232648e-06, + "loss": 0.9039, + "step": 4692 + }, + { + "epoch": 0.5642998857692539, + "grad_norm": 1.9002490407456751, + "learning_rate": 1.6819772869215972e-06, + "loss": 0.9922, + "step": 4693 + }, + { + "epoch": 0.564420128659893, + "grad_norm": 1.8781899337841645, + "learning_rate": 1.6812082504637228e-06, + "loss": 1.0533, + "step": 4694 + }, + { + "epoch": 0.564540371550532, + "grad_norm": 2.4563383207870597, + "learning_rate": 1.6804392623663025e-06, + "loss": 0.977, + "step": 4695 + }, + { + "epoch": 0.5646606144411712, + "grad_norm": 2.5132088757198088, + "learning_rate": 1.6796703227459935e-06, + "loss": 1.0188, + "step": 4696 + }, + { + "epoch": 0.5647808573318103, + "grad_norm": 2.431684746399414, + "learning_rate": 1.6789014317194407e-06, + "loss": 0.9966, + "step": 4697 + }, + { + "epoch": 0.5649011002224493, + "grad_norm": 2.9146469389214973, + "learning_rate": 1.6781325894032853e-06, + "loss": 0.9629, + "step": 4698 + }, + { + "epoch": 0.5650213431130885, + "grad_norm": 2.204894525336504, + "learning_rate": 1.6773637959141608e-06, + "loss": 1.1572, + "step": 4699 + }, + { + "epoch": 0.5651415860037275, + "grad_norm": 2.1368974963083756, + "learning_rate": 1.6765950513686915e-06, + "loss": 0.9046, + "step": 4700 + }, + { + "epoch": 0.5652618288943666, + "grad_norm": 1.7848068329396891, + "learning_rate": 1.675826355883496e-06, + "loss": 0.9971, + "step": 4701 + }, + { + "epoch": 0.5653820717850057, + "grad_norm": 1.859971728058751, + "learning_rate": 1.6750577095751848e-06, + "loss": 1.028, + "step": 4702 + }, + { + "epoch": 0.5655023146756448, + "grad_norm": 1.7263308408510125, + "learning_rate": 1.6742891125603605e-06, + "loss": 0.9656, + "step": 4703 + }, + { + "epoch": 0.5656225575662839, + "grad_norm": 2.676169173807661, + "learning_rate": 1.6735205649556185e-06, + "loss": 0.955, + "step": 4704 + }, + { + "epoch": 0.5657428004569229, + "grad_norm": 1.6386324453782009, + "learning_rate": 1.6727520668775476e-06, + "loss": 1.0857, + "step": 4705 + }, + { + "epoch": 0.5658630433475621, + "grad_norm": 1.8151349612834138, + "learning_rate": 1.6719836184427275e-06, + "loss": 0.9901, + "step": 4706 + }, + { + "epoch": 0.5659832862382012, + "grad_norm": 1.859382437041981, + "learning_rate": 1.671215219767733e-06, + "loss": 0.8791, + "step": 4707 + }, + { + "epoch": 0.5661035291288402, + "grad_norm": 2.12092007932398, + "learning_rate": 1.670446870969127e-06, + "loss": 0.9963, + "step": 4708 + }, + { + "epoch": 0.5662237720194794, + "grad_norm": 2.590943990286788, + "learning_rate": 1.6696785721634685e-06, + "loss": 1.0362, + "step": 4709 + }, + { + "epoch": 0.5663440149101184, + "grad_norm": 2.0033904186923355, + "learning_rate": 1.6689103234673086e-06, + "loss": 0.9754, + "step": 4710 + }, + { + "epoch": 0.5664642578007575, + "grad_norm": 2.1425534214762054, + "learning_rate": 1.668142124997189e-06, + "loss": 1.0041, + "step": 4711 + }, + { + "epoch": 0.5665845006913967, + "grad_norm": 0.7331154756179378, + "learning_rate": 1.6673739768696453e-06, + "loss": 0.8361, + "step": 4712 + }, + { + "epoch": 0.5667047435820357, + "grad_norm": 1.789549182511223, + "learning_rate": 1.6666058792012052e-06, + "loss": 1.011, + "step": 4713 + }, + { + "epoch": 0.5668249864726748, + "grad_norm": 0.8931250701541219, + "learning_rate": 1.6658378321083878e-06, + "loss": 0.9248, + "step": 4714 + }, + { + "epoch": 0.5669452293633139, + "grad_norm": 3.516650783508976, + "learning_rate": 1.6650698357077055e-06, + "loss": 1.0593, + "step": 4715 + }, + { + "epoch": 0.567065472253953, + "grad_norm": 11.34543397452528, + "learning_rate": 1.6643018901156632e-06, + "loss": 1.0459, + "step": 4716 + }, + { + "epoch": 0.567185715144592, + "grad_norm": 5.306305628688271, + "learning_rate": 1.6635339954487566e-06, + "loss": 1.0286, + "step": 4717 + }, + { + "epoch": 0.5673059580352312, + "grad_norm": 1.8566035926122486, + "learning_rate": 1.6627661518234765e-06, + "loss": 1.0583, + "step": 4718 + }, + { + "epoch": 0.5674262009258703, + "grad_norm": 2.7975662485414046, + "learning_rate": 1.661998359356302e-06, + "loss": 1.0884, + "step": 4719 + }, + { + "epoch": 0.5675464438165093, + "grad_norm": 0.7577081982124159, + "learning_rate": 1.6612306181637077e-06, + "loss": 0.7948, + "step": 4720 + }, + { + "epoch": 0.5676666867071485, + "grad_norm": 2.4373312671654013, + "learning_rate": 1.6604629283621598e-06, + "loss": 0.891, + "step": 4721 + }, + { + "epoch": 0.5677869295977875, + "grad_norm": 1.959351646856071, + "learning_rate": 1.6596952900681152e-06, + "loss": 0.9831, + "step": 4722 + }, + { + "epoch": 0.5679071724884266, + "grad_norm": 2.6475083063011517, + "learning_rate": 1.658927703398025e-06, + "loss": 1.0541, + "step": 4723 + }, + { + "epoch": 0.5680274153790658, + "grad_norm": 2.4756233046982605, + "learning_rate": 1.6581601684683309e-06, + "loss": 1.0137, + "step": 4724 + }, + { + "epoch": 0.5681476582697048, + "grad_norm": 2.5740982513705144, + "learning_rate": 1.6573926853954674e-06, + "loss": 0.9192, + "step": 4725 + }, + { + "epoch": 0.5682679011603439, + "grad_norm": 1.8050262001228357, + "learning_rate": 1.6566252542958608e-06, + "loss": 1.0646, + "step": 4726 + }, + { + "epoch": 0.568388144050983, + "grad_norm": 2.3178362750340606, + "learning_rate": 1.6558578752859305e-06, + "loss": 1.0221, + "step": 4727 + }, + { + "epoch": 0.5685083869416221, + "grad_norm": 2.0995330654737443, + "learning_rate": 1.6550905484820865e-06, + "loss": 1.0247, + "step": 4728 + }, + { + "epoch": 0.5686286298322611, + "grad_norm": 2.2945886767756782, + "learning_rate": 1.6543232740007328e-06, + "loss": 1.026, + "step": 4729 + }, + { + "epoch": 0.5687488727229003, + "grad_norm": 5.100579673884485, + "learning_rate": 1.653556051958263e-06, + "loss": 0.9068, + "step": 4730 + }, + { + "epoch": 0.5688691156135394, + "grad_norm": 2.475636594957541, + "learning_rate": 1.6527888824710642e-06, + "loss": 0.9751, + "step": 4731 + }, + { + "epoch": 0.5689893585041784, + "grad_norm": 2.4706678063804657, + "learning_rate": 1.6520217656555166e-06, + "loss": 0.999, + "step": 4732 + }, + { + "epoch": 0.5691096013948175, + "grad_norm": 1.7306083123479508, + "learning_rate": 1.65125470162799e-06, + "loss": 0.9457, + "step": 4733 + }, + { + "epoch": 0.5692298442854566, + "grad_norm": 3.617323013963989, + "learning_rate": 1.6504876905048485e-06, + "loss": 0.9344, + "step": 4734 + }, + { + "epoch": 0.5693500871760957, + "grad_norm": 2.19318229524353, + "learning_rate": 1.6497207324024464e-06, + "loss": 0.9606, + "step": 4735 + }, + { + "epoch": 0.5694703300667348, + "grad_norm": 3.519844202148365, + "learning_rate": 1.6489538274371305e-06, + "loss": 1.0654, + "step": 4736 + }, + { + "epoch": 0.5695905729573739, + "grad_norm": 2.4120823908852493, + "learning_rate": 1.6481869757252396e-06, + "loss": 1.0681, + "step": 4737 + }, + { + "epoch": 0.569710815848013, + "grad_norm": 1.3969423623452042, + "learning_rate": 1.647420177383105e-06, + "loss": 0.9539, + "step": 4738 + }, + { + "epoch": 0.569831058738652, + "grad_norm": 1.8510639791122416, + "learning_rate": 1.646653432527049e-06, + "loss": 0.9611, + "step": 4739 + }, + { + "epoch": 0.5699513016292912, + "grad_norm": 1.6372579264615672, + "learning_rate": 1.645886741273387e-06, + "loss": 0.9791, + "step": 4740 + }, + { + "epoch": 0.5700715445199303, + "grad_norm": 2.0082028970530894, + "learning_rate": 1.645120103738424e-06, + "loss": 0.9724, + "step": 4741 + }, + { + "epoch": 0.5701917874105693, + "grad_norm": 2.234980787838964, + "learning_rate": 1.6443535200384591e-06, + "loss": 1.0761, + "step": 4742 + }, + { + "epoch": 0.5703120303012085, + "grad_norm": 1.9758330797280812, + "learning_rate": 1.6435869902897827e-06, + "loss": 0.941, + "step": 4743 + }, + { + "epoch": 0.5704322731918475, + "grad_norm": 0.8019942609862282, + "learning_rate": 1.6428205146086764e-06, + "loss": 0.8582, + "step": 4744 + }, + { + "epoch": 0.5705525160824866, + "grad_norm": 1.660637855235956, + "learning_rate": 1.6420540931114142e-06, + "loss": 0.943, + "step": 4745 + }, + { + "epoch": 0.5706727589731257, + "grad_norm": 2.276065736086142, + "learning_rate": 1.6412877259142616e-06, + "loss": 1.0287, + "step": 4746 + }, + { + "epoch": 0.5707930018637648, + "grad_norm": 2.2179809097005094, + "learning_rate": 1.6405214131334757e-06, + "loss": 0.9775, + "step": 4747 + }, + { + "epoch": 0.5709132447544039, + "grad_norm": 1.917611766775742, + "learning_rate": 1.6397551548853052e-06, + "loss": 1.0319, + "step": 4748 + }, + { + "epoch": 0.571033487645043, + "grad_norm": 1.6692322137881157, + "learning_rate": 1.6389889512859917e-06, + "loss": 0.9461, + "step": 4749 + }, + { + "epoch": 0.5711537305356821, + "grad_norm": 0.8247590666865475, + "learning_rate": 1.638222802451767e-06, + "loss": 0.8413, + "step": 4750 + }, + { + "epoch": 0.5712739734263211, + "grad_norm": 1.6670105658815917, + "learning_rate": 1.6374567084988561e-06, + "loss": 0.9884, + "step": 4751 + }, + { + "epoch": 0.5713942163169603, + "grad_norm": 1.9633163066486237, + "learning_rate": 1.6366906695434738e-06, + "loss": 0.9995, + "step": 4752 + }, + { + "epoch": 0.5715144592075994, + "grad_norm": 2.1000726233140985, + "learning_rate": 1.6359246857018275e-06, + "loss": 1.0929, + "step": 4753 + }, + { + "epoch": 0.5716347020982384, + "grad_norm": 1.9738119535767016, + "learning_rate": 1.6351587570901178e-06, + "loss": 1.0177, + "step": 4754 + }, + { + "epoch": 0.5717549449888776, + "grad_norm": 3.315380930928313, + "learning_rate": 1.634392883824534e-06, + "loss": 0.9984, + "step": 4755 + }, + { + "epoch": 0.5718751878795166, + "grad_norm": 2.2955240441693525, + "learning_rate": 1.6336270660212595e-06, + "loss": 0.9152, + "step": 4756 + }, + { + "epoch": 0.5719954307701557, + "grad_norm": 2.2695228458321157, + "learning_rate": 1.6328613037964676e-06, + "loss": 0.8998, + "step": 4757 + }, + { + "epoch": 0.5721156736607949, + "grad_norm": 1.8481407256194304, + "learning_rate": 1.6320955972663241e-06, + "loss": 0.9173, + "step": 4758 + }, + { + "epoch": 0.5722359165514339, + "grad_norm": 1.829145423864291, + "learning_rate": 1.6313299465469857e-06, + "loss": 0.8897, + "step": 4759 + }, + { + "epoch": 0.572356159442073, + "grad_norm": 2.691746661136229, + "learning_rate": 1.6305643517546014e-06, + "loss": 1.0316, + "step": 4760 + }, + { + "epoch": 0.5724764023327121, + "grad_norm": 1.8750510526700384, + "learning_rate": 1.629798813005311e-06, + "loss": 1.0789, + "step": 4761 + }, + { + "epoch": 0.5725966452233512, + "grad_norm": 2.023652643379572, + "learning_rate": 1.6290333304152473e-06, + "loss": 0.9443, + "step": 4762 + }, + { + "epoch": 0.5727168881139902, + "grad_norm": 2.0386933325043772, + "learning_rate": 1.6282679041005314e-06, + "loss": 0.8058, + "step": 4763 + }, + { + "epoch": 0.5728371310046293, + "grad_norm": 2.286387472133604, + "learning_rate": 1.6275025341772789e-06, + "loss": 1.1116, + "step": 4764 + }, + { + "epoch": 0.5729573738952685, + "grad_norm": 2.7176096486813925, + "learning_rate": 1.626737220761596e-06, + "loss": 1.0588, + "step": 4765 + }, + { + "epoch": 0.5730776167859075, + "grad_norm": 2.151010422252622, + "learning_rate": 1.62597196396958e-06, + "loss": 1.0219, + "step": 4766 + }, + { + "epoch": 0.5731978596765466, + "grad_norm": 1.9511929530932623, + "learning_rate": 1.6252067639173197e-06, + "loss": 1.0947, + "step": 4767 + }, + { + "epoch": 0.5733181025671857, + "grad_norm": 2.016860346780795, + "learning_rate": 1.6244416207208956e-06, + "loss": 0.9381, + "step": 4768 + }, + { + "epoch": 0.5734383454578248, + "grad_norm": 1.7676769924438906, + "learning_rate": 1.6236765344963787e-06, + "loss": 0.9744, + "step": 4769 + }, + { + "epoch": 0.5735585883484638, + "grad_norm": 5.135963807997536, + "learning_rate": 1.6229115053598322e-06, + "loss": 0.9297, + "step": 4770 + }, + { + "epoch": 0.573678831239103, + "grad_norm": 1.853806821501451, + "learning_rate": 1.6221465334273108e-06, + "loss": 0.9572, + "step": 4771 + }, + { + "epoch": 0.5737990741297421, + "grad_norm": 2.352063806777691, + "learning_rate": 1.6213816188148593e-06, + "loss": 0.8516, + "step": 4772 + }, + { + "epoch": 0.5739193170203811, + "grad_norm": 1.8025531781059616, + "learning_rate": 1.6206167616385162e-06, + "loss": 1.008, + "step": 4773 + }, + { + "epoch": 0.5740395599110203, + "grad_norm": 1.9513674953942681, + "learning_rate": 1.6198519620143078e-06, + "loss": 0.9721, + "step": 4774 + }, + { + "epoch": 0.5741598028016593, + "grad_norm": 1.6317994840192627, + "learning_rate": 1.6190872200582546e-06, + "loss": 1.0166, + "step": 4775 + }, + { + "epoch": 0.5742800456922984, + "grad_norm": 10.65860451558864, + "learning_rate": 1.6183225358863676e-06, + "loss": 1.0157, + "step": 4776 + }, + { + "epoch": 0.5744002885829376, + "grad_norm": 2.360259338774185, + "learning_rate": 1.617557909614648e-06, + "loss": 0.9585, + "step": 4777 + }, + { + "epoch": 0.5745205314735766, + "grad_norm": 1.907575177952312, + "learning_rate": 1.6167933413590899e-06, + "loss": 1.0992, + "step": 4778 + }, + { + "epoch": 0.5746407743642157, + "grad_norm": 2.661358577005893, + "learning_rate": 1.6160288312356773e-06, + "loss": 1.1437, + "step": 4779 + }, + { + "epoch": 0.5747610172548548, + "grad_norm": 1.6206155462316607, + "learning_rate": 1.6152643793603857e-06, + "loss": 1.0527, + "step": 4780 + }, + { + "epoch": 0.5748812601454939, + "grad_norm": 1.681355061137822, + "learning_rate": 1.6144999858491815e-06, + "loss": 1.1153, + "step": 4781 + }, + { + "epoch": 0.575001503036133, + "grad_norm": 1.7792732915561122, + "learning_rate": 1.6137356508180232e-06, + "loss": 1.094, + "step": 4782 + }, + { + "epoch": 0.5751217459267721, + "grad_norm": 1.7804338693654402, + "learning_rate": 1.6129713743828593e-06, + "loss": 1.0507, + "step": 4783 + }, + { + "epoch": 0.5752419888174112, + "grad_norm": 1.5062366215915044, + "learning_rate": 1.6122071566596306e-06, + "loss": 0.9893, + "step": 4784 + }, + { + "epoch": 0.5753622317080502, + "grad_norm": 2.163741435668145, + "learning_rate": 1.6114429977642674e-06, + "loss": 1.0715, + "step": 4785 + }, + { + "epoch": 0.5754824745986894, + "grad_norm": 2.193131527568923, + "learning_rate": 1.6106788978126926e-06, + "loss": 0.973, + "step": 4786 + }, + { + "epoch": 0.5756027174893285, + "grad_norm": 2.4272913263833797, + "learning_rate": 1.6099148569208196e-06, + "loss": 1.029, + "step": 4787 + }, + { + "epoch": 0.5757229603799675, + "grad_norm": 1.6754715640012317, + "learning_rate": 1.6091508752045523e-06, + "loss": 0.8677, + "step": 4788 + }, + { + "epoch": 0.5758432032706067, + "grad_norm": 1.6633232516338066, + "learning_rate": 1.608386952779787e-06, + "loss": 1.1028, + "step": 4789 + }, + { + "epoch": 0.5759634461612457, + "grad_norm": 1.9839939742842814, + "learning_rate": 1.6076230897624098e-06, + "loss": 0.9811, + "step": 4790 + }, + { + "epoch": 0.5760836890518848, + "grad_norm": 2.1378806684287612, + "learning_rate": 1.6068592862682974e-06, + "loss": 1.0067, + "step": 4791 + }, + { + "epoch": 0.576203931942524, + "grad_norm": 2.2539310024231183, + "learning_rate": 1.6060955424133187e-06, + "loss": 0.971, + "step": 4792 + }, + { + "epoch": 0.576324174833163, + "grad_norm": 1.970616560850454, + "learning_rate": 1.6053318583133332e-06, + "loss": 1.1318, + "step": 4793 + }, + { + "epoch": 0.5764444177238021, + "grad_norm": 5.737353141417734, + "learning_rate": 1.6045682340841907e-06, + "loss": 0.9896, + "step": 4794 + }, + { + "epoch": 0.5765646606144411, + "grad_norm": 0.7583916800623052, + "learning_rate": 1.6038046698417336e-06, + "loss": 0.8171, + "step": 4795 + }, + { + "epoch": 0.5766849035050803, + "grad_norm": 2.550565514781071, + "learning_rate": 1.6030411657017919e-06, + "loss": 0.9251, + "step": 4796 + }, + { + "epoch": 0.5768051463957193, + "grad_norm": 1.732116947626377, + "learning_rate": 1.6022777217801903e-06, + "loss": 1.0791, + "step": 4797 + }, + { + "epoch": 0.5769253892863584, + "grad_norm": 2.576990976325961, + "learning_rate": 1.601514338192742e-06, + "loss": 0.9717, + "step": 4798 + }, + { + "epoch": 0.5770456321769976, + "grad_norm": 2.364068018572128, + "learning_rate": 1.6007510150552514e-06, + "loss": 0.9583, + "step": 4799 + }, + { + "epoch": 0.5771658750676366, + "grad_norm": 1.7300861013077449, + "learning_rate": 1.599987752483515e-06, + "loss": 0.8561, + "step": 4800 + }, + { + "epoch": 0.5772861179582757, + "grad_norm": 1.904471396249705, + "learning_rate": 1.5992245505933184e-06, + "loss": 0.9187, + "step": 4801 + }, + { + "epoch": 0.5774063608489148, + "grad_norm": 2.0399446296657295, + "learning_rate": 1.5984614095004388e-06, + "loss": 0.9488, + "step": 4802 + }, + { + "epoch": 0.5775266037395539, + "grad_norm": 2.1347629445888936, + "learning_rate": 1.5976983293206438e-06, + "loss": 1.042, + "step": 4803 + }, + { + "epoch": 0.577646846630193, + "grad_norm": 1.7908998075454985, + "learning_rate": 1.5969353101696928e-06, + "loss": 0.9509, + "step": 4804 + }, + { + "epoch": 0.5777670895208321, + "grad_norm": 1.7158067645176927, + "learning_rate": 1.5961723521633341e-06, + "loss": 1.0342, + "step": 4805 + }, + { + "epoch": 0.5778873324114712, + "grad_norm": 2.041078469801015, + "learning_rate": 1.5954094554173097e-06, + "loss": 1.1472, + "step": 4806 + }, + { + "epoch": 0.5780075753021102, + "grad_norm": 2.263887357288719, + "learning_rate": 1.5946466200473482e-06, + "loss": 1.0293, + "step": 4807 + }, + { + "epoch": 0.5781278181927494, + "grad_norm": 1.9979934163180795, + "learning_rate": 1.5938838461691723e-06, + "loss": 1.0692, + "step": 4808 + }, + { + "epoch": 0.5782480610833884, + "grad_norm": 2.727870944189456, + "learning_rate": 1.593121133898494e-06, + "loss": 1.0666, + "step": 4809 + }, + { + "epoch": 0.5783683039740275, + "grad_norm": 2.564582164856071, + "learning_rate": 1.592358483351016e-06, + "loss": 1.0287, + "step": 4810 + }, + { + "epoch": 0.5784885468646667, + "grad_norm": 1.9633872243785186, + "learning_rate": 1.5915958946424326e-06, + "loss": 0.9616, + "step": 4811 + }, + { + "epoch": 0.5786087897553057, + "grad_norm": 1.6286882945365875, + "learning_rate": 1.5908333678884271e-06, + "loss": 0.9832, + "step": 4812 + }, + { + "epoch": 0.5787290326459448, + "grad_norm": 3.4471549645836226, + "learning_rate": 1.5900709032046743e-06, + "loss": 0.9772, + "step": 4813 + }, + { + "epoch": 0.5788492755365839, + "grad_norm": 2.070764801713046, + "learning_rate": 1.5893085007068391e-06, + "loss": 1.0235, + "step": 4814 + }, + { + "epoch": 0.578969518427223, + "grad_norm": 2.0306161918695667, + "learning_rate": 1.5885461605105786e-06, + "loss": 0.9418, + "step": 4815 + }, + { + "epoch": 0.579089761317862, + "grad_norm": 2.2581652725391486, + "learning_rate": 1.5877838827315375e-06, + "loss": 1.0029, + "step": 4816 + }, + { + "epoch": 0.5792100042085012, + "grad_norm": 1.9329115155329493, + "learning_rate": 1.587021667485355e-06, + "loss": 0.9396, + "step": 4817 + }, + { + "epoch": 0.5793302470991403, + "grad_norm": 1.711316698110639, + "learning_rate": 1.5862595148876559e-06, + "loss": 1.0224, + "step": 4818 + }, + { + "epoch": 0.5794504899897793, + "grad_norm": 2.819130604966518, + "learning_rate": 1.58549742505406e-06, + "loss": 0.9996, + "step": 4819 + }, + { + "epoch": 0.5795707328804185, + "grad_norm": 2.3646977490767243, + "learning_rate": 1.5847353981001747e-06, + "loss": 0.9952, + "step": 4820 + }, + { + "epoch": 0.5796909757710575, + "grad_norm": 1.629241323427186, + "learning_rate": 1.5839734341415993e-06, + "loss": 0.9353, + "step": 4821 + }, + { + "epoch": 0.5798112186616966, + "grad_norm": 2.9778378775527377, + "learning_rate": 1.5832115332939238e-06, + "loss": 1.0035, + "step": 4822 + }, + { + "epoch": 0.5799314615523358, + "grad_norm": 1.8994262381116829, + "learning_rate": 1.5824496956727272e-06, + "loss": 0.9828, + "step": 4823 + }, + { + "epoch": 0.5800517044429748, + "grad_norm": 1.69177502847261, + "learning_rate": 1.5816879213935797e-06, + "loss": 0.9702, + "step": 4824 + }, + { + "epoch": 0.5801719473336139, + "grad_norm": 1.5796220684744264, + "learning_rate": 1.5809262105720416e-06, + "loss": 1.0308, + "step": 4825 + }, + { + "epoch": 0.580292190224253, + "grad_norm": 1.618170619069585, + "learning_rate": 1.5801645633236644e-06, + "loss": 1.0335, + "step": 4826 + }, + { + "epoch": 0.5804124331148921, + "grad_norm": 2.28462168644014, + "learning_rate": 1.579402979763989e-06, + "loss": 1.0097, + "step": 4827 + }, + { + "epoch": 0.5805326760055312, + "grad_norm": 2.050167551404169, + "learning_rate": 1.578641460008548e-06, + "loss": 1.0415, + "step": 4828 + }, + { + "epoch": 0.5806529188961702, + "grad_norm": 2.0547662872177628, + "learning_rate": 1.5778800041728613e-06, + "loss": 0.9189, + "step": 4829 + }, + { + "epoch": 0.5807731617868094, + "grad_norm": 4.55985967453658, + "learning_rate": 1.577118612372443e-06, + "loss": 0.8995, + "step": 4830 + }, + { + "epoch": 0.5808934046774484, + "grad_norm": 1.7177494951740713, + "learning_rate": 1.5763572847227943e-06, + "loss": 0.9394, + "step": 4831 + }, + { + "epoch": 0.5810136475680875, + "grad_norm": 1.8206798526608157, + "learning_rate": 1.5755960213394091e-06, + "loss": 1.0477, + "step": 4832 + }, + { + "epoch": 0.5811338904587267, + "grad_norm": 1.9226943339730551, + "learning_rate": 1.5748348223377703e-06, + "loss": 1.0176, + "step": 4833 + }, + { + "epoch": 0.5812541333493657, + "grad_norm": 2.400709536096127, + "learning_rate": 1.5740736878333507e-06, + "loss": 1.0172, + "step": 4834 + }, + { + "epoch": 0.5813743762400048, + "grad_norm": 2.8370318862688264, + "learning_rate": 1.5733126179416143e-06, + "loss": 1.0187, + "step": 4835 + }, + { + "epoch": 0.5814946191306439, + "grad_norm": 2.6098107071265666, + "learning_rate": 1.5725516127780137e-06, + "loss": 0.9625, + "step": 4836 + }, + { + "epoch": 0.581614862021283, + "grad_norm": 2.216214732896966, + "learning_rate": 1.5717906724579943e-06, + "loss": 1.1177, + "step": 4837 + }, + { + "epoch": 0.581735104911922, + "grad_norm": 2.1076137395283863, + "learning_rate": 1.571029797096989e-06, + "loss": 0.9192, + "step": 4838 + }, + { + "epoch": 0.5818553478025612, + "grad_norm": 1.9621594030194556, + "learning_rate": 1.570268986810423e-06, + "loss": 1.0256, + "step": 4839 + }, + { + "epoch": 0.5819755906932003, + "grad_norm": 11.65151742376791, + "learning_rate": 1.5695082417137096e-06, + "loss": 0.9865, + "step": 4840 + }, + { + "epoch": 0.5820958335838393, + "grad_norm": 1.5683443251632367, + "learning_rate": 1.5687475619222539e-06, + "loss": 0.9904, + "step": 4841 + }, + { + "epoch": 0.5822160764744785, + "grad_norm": 2.418216230456584, + "learning_rate": 1.5679869475514496e-06, + "loss": 0.9726, + "step": 4842 + }, + { + "epoch": 0.5823363193651175, + "grad_norm": 2.0273564509175857, + "learning_rate": 1.567226398716682e-06, + "loss": 1.0492, + "step": 4843 + }, + { + "epoch": 0.5824565622557566, + "grad_norm": 1.7895181400501996, + "learning_rate": 1.566465915533326e-06, + "loss": 0.8575, + "step": 4844 + }, + { + "epoch": 0.5825768051463958, + "grad_norm": 2.0142637875802243, + "learning_rate": 1.5657054981167458e-06, + "loss": 1.12, + "step": 4845 + }, + { + "epoch": 0.5826970480370348, + "grad_norm": 2.4681203497936792, + "learning_rate": 1.5649451465822965e-06, + "loss": 0.9128, + "step": 4846 + }, + { + "epoch": 0.5828172909276739, + "grad_norm": 1.7110681353301116, + "learning_rate": 1.5641848610453218e-06, + "loss": 1.0767, + "step": 4847 + }, + { + "epoch": 0.582937533818313, + "grad_norm": 2.034623967888009, + "learning_rate": 1.563424641621158e-06, + "loss": 1.0987, + "step": 4848 + }, + { + "epoch": 0.5830577767089521, + "grad_norm": 2.2566611974622677, + "learning_rate": 1.5626644884251282e-06, + "loss": 0.9357, + "step": 4849 + }, + { + "epoch": 0.5831780195995911, + "grad_norm": 1.6678874154551155, + "learning_rate": 1.5619044015725488e-06, + "loss": 1.1169, + "step": 4850 + }, + { + "epoch": 0.5832982624902303, + "grad_norm": 2.1424438100633036, + "learning_rate": 1.5611443811787224e-06, + "loss": 1.106, + "step": 4851 + }, + { + "epoch": 0.5834185053808694, + "grad_norm": 2.194015065234473, + "learning_rate": 1.560384427358945e-06, + "loss": 0.9315, + "step": 4852 + }, + { + "epoch": 0.5835387482715084, + "grad_norm": 1.4878778186185877, + "learning_rate": 1.5596245402284998e-06, + "loss": 0.9668, + "step": 4853 + }, + { + "epoch": 0.5836589911621476, + "grad_norm": 1.7123321005383798, + "learning_rate": 1.5588647199026619e-06, + "loss": 1.0563, + "step": 4854 + }, + { + "epoch": 0.5837792340527866, + "grad_norm": 2.420968753706024, + "learning_rate": 1.5581049664966956e-06, + "loss": 1.1135, + "step": 4855 + }, + { + "epoch": 0.5838994769434257, + "grad_norm": 1.0041814048498061, + "learning_rate": 1.5573452801258545e-06, + "loss": 0.8881, + "step": 4856 + }, + { + "epoch": 0.5840197198340649, + "grad_norm": 1.9181396035352882, + "learning_rate": 1.5565856609053824e-06, + "loss": 0.8692, + "step": 4857 + }, + { + "epoch": 0.5841399627247039, + "grad_norm": 1.8408533391848303, + "learning_rate": 1.5558261089505127e-06, + "loss": 1.0363, + "step": 4858 + }, + { + "epoch": 0.584260205615343, + "grad_norm": 2.5945796329130273, + "learning_rate": 1.5550666243764697e-06, + "loss": 1.0338, + "step": 4859 + }, + { + "epoch": 0.584380448505982, + "grad_norm": 2.218900970910436, + "learning_rate": 1.554307207298465e-06, + "loss": 1.0066, + "step": 4860 + }, + { + "epoch": 0.5845006913966212, + "grad_norm": 2.2405474576208095, + "learning_rate": 1.553547857831704e-06, + "loss": 1.0258, + "step": 4861 + }, + { + "epoch": 0.5846209342872603, + "grad_norm": 0.8958454390010986, + "learning_rate": 1.5527885760913771e-06, + "loss": 0.8792, + "step": 4862 + }, + { + "epoch": 0.5847411771778993, + "grad_norm": 1.6437183101485715, + "learning_rate": 1.552029362192668e-06, + "loss": 1.0035, + "step": 4863 + }, + { + "epoch": 0.5848614200685385, + "grad_norm": 1.8896904519311815, + "learning_rate": 1.5512702162507478e-06, + "loss": 0.9606, + "step": 4864 + }, + { + "epoch": 0.5849816629591775, + "grad_norm": 1.112120709430161, + "learning_rate": 1.5505111383807792e-06, + "loss": 0.7957, + "step": 4865 + }, + { + "epoch": 0.5851019058498166, + "grad_norm": 1.9225514159749437, + "learning_rate": 1.5497521286979138e-06, + "loss": 1.0411, + "step": 4866 + }, + { + "epoch": 0.5852221487404557, + "grad_norm": 2.4097401462246215, + "learning_rate": 1.5489931873172927e-06, + "loss": 0.983, + "step": 4867 + }, + { + "epoch": 0.5853423916310948, + "grad_norm": 1.8960580238525597, + "learning_rate": 1.5482343143540467e-06, + "loss": 1.0289, + "step": 4868 + }, + { + "epoch": 0.5854626345217339, + "grad_norm": 2.2457801666289003, + "learning_rate": 1.547475509923295e-06, + "loss": 1.0641, + "step": 4869 + }, + { + "epoch": 0.585582877412373, + "grad_norm": 0.7293923346674985, + "learning_rate": 1.5467167741401495e-06, + "loss": 0.7981, + "step": 4870 + }, + { + "epoch": 0.5857031203030121, + "grad_norm": 2.3478225928355827, + "learning_rate": 1.5459581071197083e-06, + "loss": 0.9541, + "step": 4871 + }, + { + "epoch": 0.5858233631936511, + "grad_norm": 2.2395586825647964, + "learning_rate": 1.5451995089770624e-06, + "loss": 1.0675, + "step": 4872 + }, + { + "epoch": 0.5859436060842903, + "grad_norm": 1.618090980735025, + "learning_rate": 1.5444409798272885e-06, + "loss": 0.9572, + "step": 4873 + }, + { + "epoch": 0.5860638489749294, + "grad_norm": 2.5851059236351563, + "learning_rate": 1.543682519785456e-06, + "loss": 1.0419, + "step": 4874 + }, + { + "epoch": 0.5861840918655684, + "grad_norm": 2.9374748391230447, + "learning_rate": 1.5429241289666219e-06, + "loss": 1.0375, + "step": 4875 + }, + { + "epoch": 0.5863043347562076, + "grad_norm": 2.22792113496375, + "learning_rate": 1.5421658074858342e-06, + "loss": 0.9339, + "step": 4876 + }, + { + "epoch": 0.5864245776468466, + "grad_norm": 2.3541053381389285, + "learning_rate": 1.5414075554581298e-06, + "loss": 0.9013, + "step": 4877 + }, + { + "epoch": 0.5865448205374857, + "grad_norm": 2.3476632573535707, + "learning_rate": 1.5406493729985348e-06, + "loss": 1.0221, + "step": 4878 + }, + { + "epoch": 0.5866650634281249, + "grad_norm": 3.1343503302595024, + "learning_rate": 1.5398912602220644e-06, + "loss": 0.9539, + "step": 4879 + }, + { + "epoch": 0.5867853063187639, + "grad_norm": 2.0258538977163627, + "learning_rate": 1.539133217243724e-06, + "loss": 1.026, + "step": 4880 + }, + { + "epoch": 0.586905549209403, + "grad_norm": 3.23666448232287, + "learning_rate": 1.5383752441785081e-06, + "loss": 0.9957, + "step": 4881 + }, + { + "epoch": 0.5870257921000421, + "grad_norm": 2.2347624116103337, + "learning_rate": 1.5376173411414003e-06, + "loss": 1.0907, + "step": 4882 + }, + { + "epoch": 0.5871460349906812, + "grad_norm": 1.9843930221099677, + "learning_rate": 1.5368595082473753e-06, + "loss": 1.0275, + "step": 4883 + }, + { + "epoch": 0.5872662778813202, + "grad_norm": 1.8545286989560084, + "learning_rate": 1.5361017456113935e-06, + "loss": 1.018, + "step": 4884 + }, + { + "epoch": 0.5873865207719594, + "grad_norm": 2.316653671045239, + "learning_rate": 1.5353440533484085e-06, + "loss": 1.0988, + "step": 4885 + }, + { + "epoch": 0.5875067636625985, + "grad_norm": 1.8209414725704667, + "learning_rate": 1.534586431573361e-06, + "loss": 0.8932, + "step": 4886 + }, + { + "epoch": 0.5876270065532375, + "grad_norm": 3.112824333481689, + "learning_rate": 1.5338288804011817e-06, + "loss": 1.0249, + "step": 4887 + }, + { + "epoch": 0.5877472494438767, + "grad_norm": 2.3656859214930437, + "learning_rate": 1.533071399946791e-06, + "loss": 0.9512, + "step": 4888 + }, + { + "epoch": 0.5878674923345157, + "grad_norm": 2.613411163157233, + "learning_rate": 1.5323139903250977e-06, + "loss": 0.8068, + "step": 4889 + }, + { + "epoch": 0.5879877352251548, + "grad_norm": 1.6780780267090547, + "learning_rate": 1.5315566516510002e-06, + "loss": 1.0087, + "step": 4890 + }, + { + "epoch": 0.5881079781157939, + "grad_norm": 2.18635147461206, + "learning_rate": 1.5307993840393857e-06, + "loss": 0.9142, + "step": 4891 + }, + { + "epoch": 0.588228221006433, + "grad_norm": 2.0519232776271306, + "learning_rate": 1.530042187605132e-06, + "loss": 1.0427, + "step": 4892 + }, + { + "epoch": 0.5883484638970721, + "grad_norm": 1.444172837731585, + "learning_rate": 1.5292850624631044e-06, + "loss": 1.0823, + "step": 4893 + }, + { + "epoch": 0.5884687067877111, + "grad_norm": 2.112500171548509, + "learning_rate": 1.5285280087281593e-06, + "loss": 1.0439, + "step": 4894 + }, + { + "epoch": 0.5885889496783503, + "grad_norm": 0.6772139912252775, + "learning_rate": 1.5277710265151398e-06, + "loss": 0.8039, + "step": 4895 + }, + { + "epoch": 0.5887091925689893, + "grad_norm": 2.5462200984407004, + "learning_rate": 1.5270141159388803e-06, + "loss": 1.0097, + "step": 4896 + }, + { + "epoch": 0.5888294354596284, + "grad_norm": 1.999063153191514, + "learning_rate": 1.526257277114203e-06, + "loss": 1.0414, + "step": 4897 + }, + { + "epoch": 0.5889496783502676, + "grad_norm": 1.8917400760382772, + "learning_rate": 1.5255005101559201e-06, + "loss": 1.028, + "step": 4898 + }, + { + "epoch": 0.5890699212409066, + "grad_norm": 2.3879243643330477, + "learning_rate": 1.524743815178833e-06, + "loss": 1.0065, + "step": 4899 + }, + { + "epoch": 0.5891901641315457, + "grad_norm": 2.890510185642109, + "learning_rate": 1.5239871922977315e-06, + "loss": 1.0459, + "step": 4900 + }, + { + "epoch": 0.5893104070221848, + "grad_norm": 2.0738377912723256, + "learning_rate": 1.523230641627394e-06, + "loss": 1.1297, + "step": 4901 + }, + { + "epoch": 0.5894306499128239, + "grad_norm": 2.383769659192334, + "learning_rate": 1.5224741632825888e-06, + "loss": 0.9694, + "step": 4902 + }, + { + "epoch": 0.589550892803463, + "grad_norm": 1.7916934728465328, + "learning_rate": 1.521717757378074e-06, + "loss": 0.9286, + "step": 4903 + }, + { + "epoch": 0.5896711356941021, + "grad_norm": 1.8197784950319738, + "learning_rate": 1.5209614240285943e-06, + "loss": 0.9301, + "step": 4904 + }, + { + "epoch": 0.5897913785847412, + "grad_norm": 2.0996912456785055, + "learning_rate": 1.520205163348887e-06, + "loss": 1.0827, + "step": 4905 + }, + { + "epoch": 0.5899116214753802, + "grad_norm": 0.7348769886854879, + "learning_rate": 1.519448975453674e-06, + "loss": 0.8073, + "step": 4906 + }, + { + "epoch": 0.5900318643660194, + "grad_norm": 2.073288646146913, + "learning_rate": 1.5186928604576696e-06, + "loss": 0.9951, + "step": 4907 + }, + { + "epoch": 0.5901521072566585, + "grad_norm": 3.0527042282480585, + "learning_rate": 1.5179368184755752e-06, + "loss": 1.0086, + "step": 4908 + }, + { + "epoch": 0.5902723501472975, + "grad_norm": 1.5911938982091398, + "learning_rate": 1.5171808496220821e-06, + "loss": 1.0643, + "step": 4909 + }, + { + "epoch": 0.5903925930379367, + "grad_norm": 1.8515388712602605, + "learning_rate": 1.5164249540118708e-06, + "loss": 1.0511, + "step": 4910 + }, + { + "epoch": 0.5905128359285757, + "grad_norm": 1.6482318912658693, + "learning_rate": 1.5156691317596093e-06, + "loss": 1.0736, + "step": 4911 + }, + { + "epoch": 0.5906330788192148, + "grad_norm": 2.4496576381400526, + "learning_rate": 1.5149133829799556e-06, + "loss": 0.9052, + "step": 4912 + }, + { + "epoch": 0.590753321709854, + "grad_norm": 1.845328931872589, + "learning_rate": 1.5141577077875556e-06, + "loss": 1.0419, + "step": 4913 + }, + { + "epoch": 0.590873564600493, + "grad_norm": 2.245907983146729, + "learning_rate": 1.5134021062970451e-06, + "loss": 0.9671, + "step": 4914 + }, + { + "epoch": 0.5909938074911321, + "grad_norm": 1.8604850981437715, + "learning_rate": 1.5126465786230483e-06, + "loss": 1.0451, + "step": 4915 + }, + { + "epoch": 0.5911140503817712, + "grad_norm": 2.3151871816872127, + "learning_rate": 1.5118911248801787e-06, + "loss": 1.06, + "step": 4916 + }, + { + "epoch": 0.5912342932724103, + "grad_norm": 3.247610387416352, + "learning_rate": 1.5111357451830364e-06, + "loss": 1.037, + "step": 4917 + }, + { + "epoch": 0.5913545361630493, + "grad_norm": 3.1065156143805455, + "learning_rate": 1.5103804396462131e-06, + "loss": 0.9453, + "step": 4918 + }, + { + "epoch": 0.5914747790536885, + "grad_norm": 16.33611964120123, + "learning_rate": 1.5096252083842877e-06, + "loss": 1.0396, + "step": 4919 + }, + { + "epoch": 0.5915950219443276, + "grad_norm": 1.9599259035546004, + "learning_rate": 1.5088700515118285e-06, + "loss": 1.0912, + "step": 4920 + }, + { + "epoch": 0.5917152648349666, + "grad_norm": 1.783267835798744, + "learning_rate": 1.508114969143392e-06, + "loss": 0.9023, + "step": 4921 + }, + { + "epoch": 0.5918355077256057, + "grad_norm": 1.6938703881302926, + "learning_rate": 1.5073599613935238e-06, + "loss": 1.0131, + "step": 4922 + }, + { + "epoch": 0.5919557506162448, + "grad_norm": 1.977376174811623, + "learning_rate": 1.5066050283767574e-06, + "loss": 0.8117, + "step": 4923 + }, + { + "epoch": 0.5920759935068839, + "grad_norm": 1.9683807268932092, + "learning_rate": 1.505850170207616e-06, + "loss": 1.064, + "step": 4924 + }, + { + "epoch": 0.592196236397523, + "grad_norm": 2.156687484809751, + "learning_rate": 1.505095387000611e-06, + "loss": 1.0198, + "step": 4925 + }, + { + "epoch": 0.5923164792881621, + "grad_norm": 2.0574651588651136, + "learning_rate": 1.504340678870242e-06, + "loss": 0.9792, + "step": 4926 + }, + { + "epoch": 0.5924367221788012, + "grad_norm": 2.438586359756504, + "learning_rate": 1.5035860459309989e-06, + "loss": 1.1341, + "step": 4927 + }, + { + "epoch": 0.5925569650694402, + "grad_norm": 1.852536874628057, + "learning_rate": 1.5028314882973568e-06, + "loss": 0.87, + "step": 4928 + }, + { + "epoch": 0.5926772079600794, + "grad_norm": 2.6610622123662377, + "learning_rate": 1.502077006083783e-06, + "loss": 1.0849, + "step": 4929 + }, + { + "epoch": 0.5927974508507184, + "grad_norm": 1.9768162741608308, + "learning_rate": 1.5013225994047315e-06, + "loss": 1.0038, + "step": 4930 + }, + { + "epoch": 0.5929176937413575, + "grad_norm": 1.6319516481641854, + "learning_rate": 1.5005682683746452e-06, + "loss": 1.0404, + "step": 4931 + }, + { + "epoch": 0.5930379366319967, + "grad_norm": 2.9273385765858677, + "learning_rate": 1.4998140131079553e-06, + "loss": 0.9634, + "step": 4932 + }, + { + "epoch": 0.5931581795226357, + "grad_norm": 1.9134252616295315, + "learning_rate": 1.4990598337190821e-06, + "loss": 0.9692, + "step": 4933 + }, + { + "epoch": 0.5932784224132748, + "grad_norm": 2.1457684948455364, + "learning_rate": 1.4983057303224338e-06, + "loss": 0.9166, + "step": 4934 + }, + { + "epoch": 0.5933986653039139, + "grad_norm": 4.751431701124328, + "learning_rate": 1.4975517030324072e-06, + "loss": 1.112, + "step": 4935 + }, + { + "epoch": 0.593518908194553, + "grad_norm": 0.7933272880703205, + "learning_rate": 1.4967977519633882e-06, + "loss": 0.8554, + "step": 4936 + }, + { + "epoch": 0.593639151085192, + "grad_norm": 2.204102644203877, + "learning_rate": 1.4960438772297494e-06, + "loss": 1.0192, + "step": 4937 + }, + { + "epoch": 0.5937593939758312, + "grad_norm": 6.198287653572217, + "learning_rate": 1.495290078945855e-06, + "loss": 0.9748, + "step": 4938 + }, + { + "epoch": 0.5938796368664703, + "grad_norm": 1.9214591840289719, + "learning_rate": 1.4945363572260529e-06, + "loss": 0.9817, + "step": 4939 + }, + { + "epoch": 0.5939998797571093, + "grad_norm": 2.9659629187089056, + "learning_rate": 1.4937827121846845e-06, + "loss": 0.9111, + "step": 4940 + }, + { + "epoch": 0.5941201226477485, + "grad_norm": 1.5309474412603494, + "learning_rate": 1.4930291439360755e-06, + "loss": 0.974, + "step": 4941 + }, + { + "epoch": 0.5942403655383875, + "grad_norm": 2.3617740585508593, + "learning_rate": 1.4922756525945427e-06, + "loss": 1.0309, + "step": 4942 + }, + { + "epoch": 0.5943606084290266, + "grad_norm": 0.7907318736581838, + "learning_rate": 1.4915222382743894e-06, + "loss": 0.8332, + "step": 4943 + }, + { + "epoch": 0.5944808513196658, + "grad_norm": 2.2323320559290605, + "learning_rate": 1.4907689010899085e-06, + "loss": 0.9574, + "step": 4944 + }, + { + "epoch": 0.5946010942103048, + "grad_norm": 3.9127116965647457, + "learning_rate": 1.4900156411553804e-06, + "loss": 0.8593, + "step": 4945 + }, + { + "epoch": 0.5947213371009439, + "grad_norm": 2.0482219894022946, + "learning_rate": 1.4892624585850739e-06, + "loss": 1.0877, + "step": 4946 + }, + { + "epoch": 0.594841579991583, + "grad_norm": 2.063333978524464, + "learning_rate": 1.4885093534932465e-06, + "loss": 1.0335, + "step": 4947 + }, + { + "epoch": 0.5949618228822221, + "grad_norm": 2.0947483017155744, + "learning_rate": 1.4877563259941433e-06, + "loss": 0.9505, + "step": 4948 + }, + { + "epoch": 0.5950820657728612, + "grad_norm": 3.062850309305356, + "learning_rate": 1.4870033762019988e-06, + "loss": 0.9171, + "step": 4949 + }, + { + "epoch": 0.5952023086635003, + "grad_norm": 1.6152839699157535, + "learning_rate": 1.4862505042310334e-06, + "loss": 0.9686, + "step": 4950 + }, + { + "epoch": 0.5953225515541394, + "grad_norm": 1.6553231290937267, + "learning_rate": 1.4854977101954587e-06, + "loss": 0.9359, + "step": 4951 + }, + { + "epoch": 0.5954427944447784, + "grad_norm": 2.2494341880598228, + "learning_rate": 1.4847449942094716e-06, + "loss": 1.0992, + "step": 4952 + }, + { + "epoch": 0.5955630373354175, + "grad_norm": 9.243653465731569, + "learning_rate": 1.4839923563872598e-06, + "loss": 1.1004, + "step": 4953 + }, + { + "epoch": 0.5956832802260567, + "grad_norm": 1.8204062769232376, + "learning_rate": 1.483239796842997e-06, + "loss": 0.9971, + "step": 4954 + }, + { + "epoch": 0.5958035231166957, + "grad_norm": 1.7873378253274654, + "learning_rate": 1.4824873156908462e-06, + "loss": 1.0785, + "step": 4955 + }, + { + "epoch": 0.5959237660073348, + "grad_norm": 1.9619912283776175, + "learning_rate": 1.4817349130449584e-06, + "loss": 0.9915, + "step": 4956 + }, + { + "epoch": 0.5960440088979739, + "grad_norm": 1.785595795923671, + "learning_rate": 1.4809825890194717e-06, + "loss": 1.0648, + "step": 4957 + }, + { + "epoch": 0.596164251788613, + "grad_norm": 1.869664038195808, + "learning_rate": 1.4802303437285139e-06, + "loss": 1.0071, + "step": 4958 + }, + { + "epoch": 0.596284494679252, + "grad_norm": 2.082518558121842, + "learning_rate": 1.4794781772861994e-06, + "loss": 1.0385, + "step": 4959 + }, + { + "epoch": 0.5964047375698912, + "grad_norm": 2.091158273421577, + "learning_rate": 1.4787260898066324e-06, + "loss": 0.9067, + "step": 4960 + }, + { + "epoch": 0.5965249804605303, + "grad_norm": 2.04653019221235, + "learning_rate": 1.4779740814039023e-06, + "loss": 1.0983, + "step": 4961 + }, + { + "epoch": 0.5966452233511693, + "grad_norm": 1.9776778862345228, + "learning_rate": 1.4772221521920894e-06, + "loss": 0.9182, + "step": 4962 + }, + { + "epoch": 0.5967654662418085, + "grad_norm": 2.0232841993312327, + "learning_rate": 1.4764703022852598e-06, + "loss": 0.9791, + "step": 4963 + }, + { + "epoch": 0.5968857091324475, + "grad_norm": 1.979959095041079, + "learning_rate": 1.4757185317974696e-06, + "loss": 1.0069, + "step": 4964 + }, + { + "epoch": 0.5970059520230866, + "grad_norm": 2.1998364951323457, + "learning_rate": 1.474966840842761e-06, + "loss": 0.9492, + "step": 4965 + }, + { + "epoch": 0.5971261949137258, + "grad_norm": 1.7542273008811424, + "learning_rate": 1.4742152295351655e-06, + "loss": 1.1049, + "step": 4966 + }, + { + "epoch": 0.5972464378043648, + "grad_norm": 2.450397797812264, + "learning_rate": 1.4734636979887016e-06, + "loss": 0.8707, + "step": 4967 + }, + { + "epoch": 0.5973666806950039, + "grad_norm": 2.1525597437986614, + "learning_rate": 1.4727122463173755e-06, + "loss": 1.1368, + "step": 4968 + }, + { + "epoch": 0.597486923585643, + "grad_norm": 1.8614510238438484, + "learning_rate": 1.471960874635183e-06, + "loss": 0.8778, + "step": 4969 + }, + { + "epoch": 0.5976071664762821, + "grad_norm": 2.2888751164837813, + "learning_rate": 1.4712095830561055e-06, + "loss": 0.9409, + "step": 4970 + }, + { + "epoch": 0.5977274093669211, + "grad_norm": 1.9444475204201241, + "learning_rate": 1.4704583716941147e-06, + "loss": 1.0449, + "step": 4971 + }, + { + "epoch": 0.5978476522575603, + "grad_norm": 1.9526537517425704, + "learning_rate": 1.4697072406631672e-06, + "loss": 0.9576, + "step": 4972 + }, + { + "epoch": 0.5979678951481994, + "grad_norm": 1.7232076925720539, + "learning_rate": 1.4689561900772097e-06, + "loss": 0.9622, + "step": 4973 + }, + { + "epoch": 0.5980881380388384, + "grad_norm": 2.5981560477245464, + "learning_rate": 1.4682052200501758e-06, + "loss": 0.961, + "step": 4974 + }, + { + "epoch": 0.5982083809294776, + "grad_norm": 1.863155720638875, + "learning_rate": 1.4674543306959876e-06, + "loss": 1.0352, + "step": 4975 + }, + { + "epoch": 0.5983286238201166, + "grad_norm": 3.9277738306462444, + "learning_rate": 1.4667035221285535e-06, + "loss": 1.0845, + "step": 4976 + }, + { + "epoch": 0.5984488667107557, + "grad_norm": 1.8688117588006403, + "learning_rate": 1.4659527944617715e-06, + "loss": 0.9816, + "step": 4977 + }, + { + "epoch": 0.5985691096013949, + "grad_norm": 1.7029379645722977, + "learning_rate": 1.465202147809526e-06, + "loss": 0.994, + "step": 4978 + }, + { + "epoch": 0.5986893524920339, + "grad_norm": 2.0029494471995815, + "learning_rate": 1.4644515822856888e-06, + "loss": 0.9991, + "step": 4979 + }, + { + "epoch": 0.598809595382673, + "grad_norm": 0.7668109194868065, + "learning_rate": 1.4637010980041215e-06, + "loss": 0.8031, + "step": 4980 + }, + { + "epoch": 0.5989298382733121, + "grad_norm": 2.3944836560446476, + "learning_rate": 1.4629506950786707e-06, + "loss": 1.1314, + "step": 4981 + }, + { + "epoch": 0.5990500811639512, + "grad_norm": 0.8179280457666219, + "learning_rate": 1.4622003736231733e-06, + "loss": 0.7985, + "step": 4982 + }, + { + "epoch": 0.5991703240545903, + "grad_norm": 1.994114142381235, + "learning_rate": 1.461450133751451e-06, + "loss": 1.0417, + "step": 4983 + }, + { + "epoch": 0.5992905669452293, + "grad_norm": 2.5861897417573783, + "learning_rate": 1.4606999755773153e-06, + "loss": 0.9936, + "step": 4984 + }, + { + "epoch": 0.5994108098358685, + "grad_norm": 2.3065611383888642, + "learning_rate": 1.4599498992145643e-06, + "loss": 1.0592, + "step": 4985 + }, + { + "epoch": 0.5995310527265075, + "grad_norm": 1.9234670232371, + "learning_rate": 1.4591999047769846e-06, + "loss": 0.9434, + "step": 4986 + }, + { + "epoch": 0.5996512956171466, + "grad_norm": 2.0037124272355924, + "learning_rate": 1.4584499923783486e-06, + "loss": 0.9881, + "step": 4987 + }, + { + "epoch": 0.5997715385077858, + "grad_norm": 1.9942096096738862, + "learning_rate": 1.457700162132419e-06, + "loss": 0.9975, + "step": 4988 + }, + { + "epoch": 0.5998917813984248, + "grad_norm": 1.9535876527233604, + "learning_rate": 1.4569504141529433e-06, + "loss": 0.9606, + "step": 4989 + }, + { + "epoch": 0.6000120242890639, + "grad_norm": 3.5327400507156166, + "learning_rate": 1.456200748553658e-06, + "loss": 0.9561, + "step": 4990 + }, + { + "epoch": 0.600132267179703, + "grad_norm": 1.6462418093582578, + "learning_rate": 1.455451165448287e-06, + "loss": 1.0242, + "step": 4991 + }, + { + "epoch": 0.6002525100703421, + "grad_norm": 2.506394695566642, + "learning_rate": 1.4547016649505407e-06, + "loss": 0.9725, + "step": 4992 + }, + { + "epoch": 0.6003727529609811, + "grad_norm": 2.115601906586998, + "learning_rate": 1.4539522471741193e-06, + "loss": 1.0862, + "step": 4993 + }, + { + "epoch": 0.6004929958516203, + "grad_norm": 2.4290544967097465, + "learning_rate": 1.4532029122327067e-06, + "loss": 0.9493, + "step": 4994 + }, + { + "epoch": 0.6006132387422594, + "grad_norm": 2.0513208460501593, + "learning_rate": 1.4524536602399783e-06, + "loss": 0.9968, + "step": 4995 + }, + { + "epoch": 0.6007334816328984, + "grad_norm": 1.623912080306326, + "learning_rate": 1.4517044913095938e-06, + "loss": 1.011, + "step": 4996 + }, + { + "epoch": 0.6008537245235376, + "grad_norm": 1.9024564173113894, + "learning_rate": 1.4509554055552022e-06, + "loss": 1.054, + "step": 4997 + }, + { + "epoch": 0.6009739674141766, + "grad_norm": 2.3638030681977185, + "learning_rate": 1.450206403090439e-06, + "loss": 1.0781, + "step": 4998 + }, + { + "epoch": 0.6010942103048157, + "grad_norm": 3.826175987627001, + "learning_rate": 1.4494574840289274e-06, + "loss": 1.0998, + "step": 4999 + }, + { + "epoch": 0.6012144531954549, + "grad_norm": 1.6785694748423607, + "learning_rate": 1.4487086484842782e-06, + "loss": 0.9753, + "step": 5000 + }, + { + "epoch": 0.6013346960860939, + "grad_norm": 4.3867091843729185, + "learning_rate": 1.4479598965700878e-06, + "loss": 0.8406, + "step": 5001 + }, + { + "epoch": 0.601454938976733, + "grad_norm": 2.968106812521596, + "learning_rate": 1.4472112283999427e-06, + "loss": 0.9257, + "step": 5002 + }, + { + "epoch": 0.6015751818673721, + "grad_norm": 2.3738228741829888, + "learning_rate": 1.4464626440874143e-06, + "loss": 0.9266, + "step": 5003 + }, + { + "epoch": 0.6016954247580112, + "grad_norm": 2.544046153231688, + "learning_rate": 1.4457141437460636e-06, + "loss": 0.98, + "step": 5004 + }, + { + "epoch": 0.6018156676486502, + "grad_norm": 1.7628742497040948, + "learning_rate": 1.444965727489436e-06, + "loss": 0.968, + "step": 5005 + }, + { + "epoch": 0.6019359105392894, + "grad_norm": 9.840049605205767, + "learning_rate": 1.444217395431066e-06, + "loss": 0.8688, + "step": 5006 + }, + { + "epoch": 0.6020561534299285, + "grad_norm": 0.8007202497369326, + "learning_rate": 1.4434691476844755e-06, + "loss": 0.7959, + "step": 5007 + }, + { + "epoch": 0.6021763963205675, + "grad_norm": 2.43531867994858, + "learning_rate": 1.4427209843631729e-06, + "loss": 0.9073, + "step": 5008 + }, + { + "epoch": 0.6022966392112067, + "grad_norm": 1.835688147960459, + "learning_rate": 1.4419729055806534e-06, + "loss": 1.0472, + "step": 5009 + }, + { + "epoch": 0.6024168821018457, + "grad_norm": 2.202076764008231, + "learning_rate": 1.441224911450401e-06, + "loss": 1.061, + "step": 5010 + }, + { + "epoch": 0.6025371249924848, + "grad_norm": 1.7102919802209307, + "learning_rate": 1.4404770020858851e-06, + "loss": 1.0583, + "step": 5011 + }, + { + "epoch": 0.602657367883124, + "grad_norm": 1.8461311610980002, + "learning_rate": 1.439729177600563e-06, + "loss": 1.1021, + "step": 5012 + }, + { + "epoch": 0.602777610773763, + "grad_norm": 3.1824051773796334, + "learning_rate": 1.4389814381078793e-06, + "loss": 0.9689, + "step": 5013 + }, + { + "epoch": 0.6028978536644021, + "grad_norm": 2.365498661109418, + "learning_rate": 1.438233783721265e-06, + "loss": 1.0376, + "step": 5014 + }, + { + "epoch": 0.6030180965550412, + "grad_norm": 2.2698031080452754, + "learning_rate": 1.43748621455414e-06, + "loss": 1.0152, + "step": 5015 + }, + { + "epoch": 0.6031383394456803, + "grad_norm": 2.512574804910006, + "learning_rate": 1.4367387307199082e-06, + "loss": 1.0498, + "step": 5016 + }, + { + "epoch": 0.6032585823363193, + "grad_norm": 2.185960282267694, + "learning_rate": 1.4359913323319632e-06, + "loss": 1.0603, + "step": 5017 + }, + { + "epoch": 0.6033788252269584, + "grad_norm": 2.0556183653620317, + "learning_rate": 1.4352440195036847e-06, + "loss": 1.0127, + "step": 5018 + }, + { + "epoch": 0.6034990681175976, + "grad_norm": 1.7082641479004572, + "learning_rate": 1.4344967923484395e-06, + "loss": 1.0352, + "step": 5019 + }, + { + "epoch": 0.6036193110082366, + "grad_norm": 2.6561901310176346, + "learning_rate": 1.433749650979581e-06, + "loss": 0.9568, + "step": 5020 + }, + { + "epoch": 0.6037395538988757, + "grad_norm": 1.921698461348418, + "learning_rate": 1.433002595510451e-06, + "loss": 0.9218, + "step": 5021 + }, + { + "epoch": 0.6038597967895148, + "grad_norm": 1.8935189915125943, + "learning_rate": 1.4322556260543757e-06, + "loss": 0.953, + "step": 5022 + }, + { + "epoch": 0.6039800396801539, + "grad_norm": 0.90566573546985, + "learning_rate": 1.4315087427246703e-06, + "loss": 0.8645, + "step": 5023 + }, + { + "epoch": 0.604100282570793, + "grad_norm": 0.8917616567788254, + "learning_rate": 1.4307619456346372e-06, + "loss": 0.8228, + "step": 5024 + }, + { + "epoch": 0.6042205254614321, + "grad_norm": 6.797671637491922, + "learning_rate": 1.430015234897564e-06, + "loss": 0.9775, + "step": 5025 + }, + { + "epoch": 0.6043407683520712, + "grad_norm": 1.9704019790890381, + "learning_rate": 1.4292686106267274e-06, + "loss": 0.902, + "step": 5026 + }, + { + "epoch": 0.6044610112427102, + "grad_norm": 1.8827219581202463, + "learning_rate": 1.4285220729353876e-06, + "loss": 1.0061, + "step": 5027 + }, + { + "epoch": 0.6045812541333494, + "grad_norm": 2.161451175785087, + "learning_rate": 1.4277756219367957e-06, + "loss": 1.0153, + "step": 5028 + }, + { + "epoch": 0.6047014970239885, + "grad_norm": 2.07621597142729, + "learning_rate": 1.4270292577441864e-06, + "loss": 1.0279, + "step": 5029 + }, + { + "epoch": 0.6048217399146275, + "grad_norm": 1.6316687123276405, + "learning_rate": 1.4262829804707836e-06, + "loss": 0.9557, + "step": 5030 + }, + { + "epoch": 0.6049419828052667, + "grad_norm": 1.6179296292632146, + "learning_rate": 1.4255367902297958e-06, + "loss": 0.9326, + "step": 5031 + }, + { + "epoch": 0.6050622256959057, + "grad_norm": 2.4154787322813496, + "learning_rate": 1.4247906871344215e-06, + "loss": 1.0218, + "step": 5032 + }, + { + "epoch": 0.6051824685865448, + "grad_norm": 11.30875735255146, + "learning_rate": 1.4240446712978415e-06, + "loss": 0.9892, + "step": 5033 + }, + { + "epoch": 0.605302711477184, + "grad_norm": 2.194876305086213, + "learning_rate": 1.423298742833227e-06, + "loss": 0.9787, + "step": 5034 + }, + { + "epoch": 0.605422954367823, + "grad_norm": 4.267403815090027, + "learning_rate": 1.4225529018537352e-06, + "loss": 0.9549, + "step": 5035 + }, + { + "epoch": 0.6055431972584621, + "grad_norm": 1.6646946525670707, + "learning_rate": 1.4218071484725082e-06, + "loss": 1.0135, + "step": 5036 + }, + { + "epoch": 0.6056634401491012, + "grad_norm": 2.1468808725157356, + "learning_rate": 1.4210614828026786e-06, + "loss": 0.9997, + "step": 5037 + }, + { + "epoch": 0.6057836830397403, + "grad_norm": 1.5836482320483527, + "learning_rate": 1.4203159049573605e-06, + "loss": 0.9819, + "step": 5038 + }, + { + "epoch": 0.6059039259303793, + "grad_norm": 2.0713642256002287, + "learning_rate": 1.4195704150496593e-06, + "loss": 1.1059, + "step": 5039 + }, + { + "epoch": 0.6060241688210185, + "grad_norm": 1.6811200797214418, + "learning_rate": 1.4188250131926639e-06, + "loss": 0.9737, + "step": 5040 + }, + { + "epoch": 0.6061444117116576, + "grad_norm": 2.1012232623648592, + "learning_rate": 1.4180796994994525e-06, + "loss": 1.0436, + "step": 5041 + }, + { + "epoch": 0.6062646546022966, + "grad_norm": 1.8367889377905524, + "learning_rate": 1.4173344740830877e-06, + "loss": 0.9549, + "step": 5042 + }, + { + "epoch": 0.6063848974929358, + "grad_norm": 1.7120664172970417, + "learning_rate": 1.4165893370566206e-06, + "loss": 0.9403, + "step": 5043 + }, + { + "epoch": 0.6065051403835748, + "grad_norm": 2.181028876259076, + "learning_rate": 1.4158442885330865e-06, + "loss": 1.0113, + "step": 5044 + }, + { + "epoch": 0.6066253832742139, + "grad_norm": 2.265533550653029, + "learning_rate": 1.4150993286255094e-06, + "loss": 1.0234, + "step": 5045 + }, + { + "epoch": 0.6067456261648531, + "grad_norm": 2.791759736137121, + "learning_rate": 1.4143544574468993e-06, + "loss": 1.0347, + "step": 5046 + }, + { + "epoch": 0.6068658690554921, + "grad_norm": 1.9513864943315047, + "learning_rate": 1.4136096751102523e-06, + "loss": 1.0611, + "step": 5047 + }, + { + "epoch": 0.6069861119461312, + "grad_norm": 1.980376292220512, + "learning_rate": 1.4128649817285516e-06, + "loss": 1.0676, + "step": 5048 + }, + { + "epoch": 0.6071063548367702, + "grad_norm": 2.145637602042646, + "learning_rate": 1.412120377414766e-06, + "loss": 0.8662, + "step": 5049 + }, + { + "epoch": 0.6072265977274094, + "grad_norm": 1.5857966318191667, + "learning_rate": 1.4113758622818522e-06, + "loss": 0.9485, + "step": 5050 + }, + { + "epoch": 0.6073468406180484, + "grad_norm": 1.8861919165099803, + "learning_rate": 1.410631436442751e-06, + "loss": 1.068, + "step": 5051 + }, + { + "epoch": 0.6074670835086875, + "grad_norm": 2.1087290587230565, + "learning_rate": 1.4098871000103936e-06, + "loss": 1.1014, + "step": 5052 + }, + { + "epoch": 0.6075873263993267, + "grad_norm": 2.439284625010276, + "learning_rate": 1.409142853097693e-06, + "loss": 1.0617, + "step": 5053 + }, + { + "epoch": 0.6077075692899657, + "grad_norm": 2.036434072624797, + "learning_rate": 1.408398695817553e-06, + "loss": 1.0357, + "step": 5054 + }, + { + "epoch": 0.6078278121806048, + "grad_norm": 2.3476876305944, + "learning_rate": 1.4076546282828593e-06, + "loss": 0.9399, + "step": 5055 + }, + { + "epoch": 0.6079480550712439, + "grad_norm": 2.4999805449682455, + "learning_rate": 1.4069106506064874e-06, + "loss": 0.8994, + "step": 5056 + }, + { + "epoch": 0.608068297961883, + "grad_norm": 1.94539841305172, + "learning_rate": 1.4061667629012989e-06, + "loss": 1.0221, + "step": 5057 + }, + { + "epoch": 0.608188540852522, + "grad_norm": 3.88166211776718, + "learning_rate": 1.40542296528014e-06, + "loss": 1.0679, + "step": 5058 + }, + { + "epoch": 0.6083087837431612, + "grad_norm": 1.9729363875190649, + "learning_rate": 1.4046792578558452e-06, + "loss": 0.9976, + "step": 5059 + }, + { + "epoch": 0.6084290266338003, + "grad_norm": 2.380817115665256, + "learning_rate": 1.4039356407412325e-06, + "loss": 0.9945, + "step": 5060 + }, + { + "epoch": 0.6085492695244393, + "grad_norm": 0.7927046086895962, + "learning_rate": 1.40319211404911e-06, + "loss": 0.8091, + "step": 5061 + }, + { + "epoch": 0.6086695124150785, + "grad_norm": 2.043021848728874, + "learning_rate": 1.4024486778922691e-06, + "loss": 1.1456, + "step": 5062 + }, + { + "epoch": 0.6087897553057176, + "grad_norm": 2.0137597258206856, + "learning_rate": 1.4017053323834884e-06, + "loss": 1.0081, + "step": 5063 + }, + { + "epoch": 0.6089099981963566, + "grad_norm": 2.6497577088631954, + "learning_rate": 1.4009620776355333e-06, + "loss": 1.0005, + "step": 5064 + }, + { + "epoch": 0.6090302410869958, + "grad_norm": 1.7963082124519485, + "learning_rate": 1.4002189137611553e-06, + "loss": 1.0295, + "step": 5065 + }, + { + "epoch": 0.6091504839776348, + "grad_norm": 1.8039793796841026, + "learning_rate": 1.3994758408730901e-06, + "loss": 0.9333, + "step": 5066 + }, + { + "epoch": 0.6092707268682739, + "grad_norm": 2.449911843874209, + "learning_rate": 1.3987328590840629e-06, + "loss": 0.9987, + "step": 5067 + }, + { + "epoch": 0.609390969758913, + "grad_norm": 1.9923139943373562, + "learning_rate": 1.397989968506783e-06, + "loss": 1.1005, + "step": 5068 + }, + { + "epoch": 0.6095112126495521, + "grad_norm": 2.421663927294168, + "learning_rate": 1.3972471692539458e-06, + "loss": 0.9614, + "step": 5069 + }, + { + "epoch": 0.6096314555401912, + "grad_norm": 2.130639950427607, + "learning_rate": 1.3965044614382348e-06, + "loss": 0.9875, + "step": 5070 + }, + { + "epoch": 0.6097516984308303, + "grad_norm": 2.4260561383412083, + "learning_rate": 1.3957618451723162e-06, + "loss": 0.9949, + "step": 5071 + }, + { + "epoch": 0.6098719413214694, + "grad_norm": 1.9978339863980268, + "learning_rate": 1.3950193205688457e-06, + "loss": 0.9533, + "step": 5072 + }, + { + "epoch": 0.6099921842121084, + "grad_norm": 1.9351739764797338, + "learning_rate": 1.3942768877404627e-06, + "loss": 1.0773, + "step": 5073 + }, + { + "epoch": 0.6101124271027476, + "grad_norm": 1.5295679624334713, + "learning_rate": 1.393534546799795e-06, + "loss": 0.9721, + "step": 5074 + }, + { + "epoch": 0.6102326699933867, + "grad_norm": 1.8399720359833416, + "learning_rate": 1.3927922978594536e-06, + "loss": 0.9193, + "step": 5075 + }, + { + "epoch": 0.6103529128840257, + "grad_norm": 0.7839346344834693, + "learning_rate": 1.3920501410320387e-06, + "loss": 0.8117, + "step": 5076 + }, + { + "epoch": 0.6104731557746649, + "grad_norm": 2.1890588927800847, + "learning_rate": 1.3913080764301333e-06, + "loss": 0.9976, + "step": 5077 + }, + { + "epoch": 0.6105933986653039, + "grad_norm": 1.880199217488037, + "learning_rate": 1.3905661041663085e-06, + "loss": 0.9533, + "step": 5078 + }, + { + "epoch": 0.610713641555943, + "grad_norm": 1.9710310785019738, + "learning_rate": 1.389824224353122e-06, + "loss": 0.889, + "step": 5079 + }, + { + "epoch": 0.610833884446582, + "grad_norm": 1.501070911864689, + "learning_rate": 1.389082437103115e-06, + "loss": 1.0059, + "step": 5080 + }, + { + "epoch": 0.6109541273372212, + "grad_norm": 1.8103599410686806, + "learning_rate": 1.3883407425288172e-06, + "loss": 1.0146, + "step": 5081 + }, + { + "epoch": 0.6110743702278603, + "grad_norm": 2.3536871250079052, + "learning_rate": 1.3875991407427417e-06, + "loss": 1.0328, + "step": 5082 + }, + { + "epoch": 0.6111946131184993, + "grad_norm": 0.7734184262784836, + "learning_rate": 1.38685763185739e-06, + "loss": 0.8197, + "step": 5083 + }, + { + "epoch": 0.6113148560091385, + "grad_norm": 2.5201435143612945, + "learning_rate": 1.3861162159852476e-06, + "loss": 0.9089, + "step": 5084 + }, + { + "epoch": 0.6114350988997775, + "grad_norm": 2.0523093497679086, + "learning_rate": 1.3853748932387875e-06, + "loss": 1.039, + "step": 5085 + }, + { + "epoch": 0.6115553417904166, + "grad_norm": 2.3642586192429436, + "learning_rate": 1.3846336637304671e-06, + "loss": 0.9858, + "step": 5086 + }, + { + "epoch": 0.6116755846810558, + "grad_norm": 5.794227674230648, + "learning_rate": 1.3838925275727316e-06, + "loss": 1.0675, + "step": 5087 + }, + { + "epoch": 0.6117958275716948, + "grad_norm": 2.067975146345357, + "learning_rate": 1.3831514848780089e-06, + "loss": 1.0283, + "step": 5088 + }, + { + "epoch": 0.6119160704623339, + "grad_norm": 2.7879178986959756, + "learning_rate": 1.3824105357587152e-06, + "loss": 1.1607, + "step": 5089 + }, + { + "epoch": 0.612036313352973, + "grad_norm": 1.7365579013238426, + "learning_rate": 1.381669680327253e-06, + "loss": 1.0621, + "step": 5090 + }, + { + "epoch": 0.6121565562436121, + "grad_norm": 1.8453169807295702, + "learning_rate": 1.380928918696008e-06, + "loss": 0.9478, + "step": 5091 + }, + { + "epoch": 0.6122767991342511, + "grad_norm": 2.351338930369761, + "learning_rate": 1.3801882509773548e-06, + "loss": 0.9531, + "step": 5092 + }, + { + "epoch": 0.6123970420248903, + "grad_norm": 1.766994316679143, + "learning_rate": 1.3794476772836503e-06, + "loss": 1.0553, + "step": 5093 + }, + { + "epoch": 0.6125172849155294, + "grad_norm": 1.7497128523529473, + "learning_rate": 1.3787071977272402e-06, + "loss": 1.0841, + "step": 5094 + }, + { + "epoch": 0.6126375278061684, + "grad_norm": 3.5822220047439792, + "learning_rate": 1.3779668124204535e-06, + "loss": 0.9557, + "step": 5095 + }, + { + "epoch": 0.6127577706968076, + "grad_norm": 1.7154193859188434, + "learning_rate": 1.3772265214756074e-06, + "loss": 1.0472, + "step": 5096 + }, + { + "epoch": 0.6128780135874466, + "grad_norm": 3.704952770760963, + "learning_rate": 1.3764863250050025e-06, + "loss": 0.9921, + "step": 5097 + }, + { + "epoch": 0.6129982564780857, + "grad_norm": 1.8159491507347334, + "learning_rate": 1.3757462231209272e-06, + "loss": 1.0429, + "step": 5098 + }, + { + "epoch": 0.6131184993687249, + "grad_norm": 2.7399338375930187, + "learning_rate": 1.3750062159356525e-06, + "loss": 1.1267, + "step": 5099 + }, + { + "epoch": 0.6132387422593639, + "grad_norm": 1.849553881802659, + "learning_rate": 1.3742663035614382e-06, + "loss": 1.0686, + "step": 5100 + }, + { + "epoch": 0.613358985150003, + "grad_norm": 1.7830249576201842, + "learning_rate": 1.3735264861105283e-06, + "loss": 1.0397, + "step": 5101 + }, + { + "epoch": 0.6134792280406421, + "grad_norm": 2.3468909007827787, + "learning_rate": 1.372786763695152e-06, + "loss": 1.024, + "step": 5102 + }, + { + "epoch": 0.6135994709312812, + "grad_norm": 2.1431644718854983, + "learning_rate": 1.3720471364275257e-06, + "loss": 1.0146, + "step": 5103 + }, + { + "epoch": 0.6137197138219203, + "grad_norm": 1.8921729442436637, + "learning_rate": 1.3713076044198486e-06, + "loss": 1.016, + "step": 5104 + }, + { + "epoch": 0.6138399567125594, + "grad_norm": 2.152767963455416, + "learning_rate": 1.3705681677843086e-06, + "loss": 1.045, + "step": 5105 + }, + { + "epoch": 0.6139601996031985, + "grad_norm": 0.7814446397552989, + "learning_rate": 1.3698288266330768e-06, + "loss": 0.8433, + "step": 5106 + }, + { + "epoch": 0.6140804424938375, + "grad_norm": 3.395469193544763, + "learning_rate": 1.3690895810783113e-06, + "loss": 0.964, + "step": 5107 + }, + { + "epoch": 0.6142006853844767, + "grad_norm": 1.9982970855751199, + "learning_rate": 1.3683504312321543e-06, + "loss": 0.95, + "step": 5108 + }, + { + "epoch": 0.6143209282751158, + "grad_norm": 5.275855384643599, + "learning_rate": 1.3676113772067355e-06, + "loss": 1.0379, + "step": 5109 + }, + { + "epoch": 0.6144411711657548, + "grad_norm": 2.073897687126513, + "learning_rate": 1.3668724191141671e-06, + "loss": 0.9634, + "step": 5110 + }, + { + "epoch": 0.6145614140563939, + "grad_norm": 2.524648653268159, + "learning_rate": 1.3661335570665493e-06, + "loss": 0.9011, + "step": 5111 + }, + { + "epoch": 0.614681656947033, + "grad_norm": 2.5899112284113186, + "learning_rate": 1.3653947911759676e-06, + "loss": 0.9318, + "step": 5112 + }, + { + "epoch": 0.6148018998376721, + "grad_norm": 2.0096178065325057, + "learning_rate": 1.3646561215544904e-06, + "loss": 0.9854, + "step": 5113 + }, + { + "epoch": 0.6149221427283111, + "grad_norm": 2.0332003106968997, + "learning_rate": 1.363917548314176e-06, + "loss": 1.03, + "step": 5114 + }, + { + "epoch": 0.6150423856189503, + "grad_norm": 1.886540690800401, + "learning_rate": 1.3631790715670626e-06, + "loss": 0.9705, + "step": 5115 + }, + { + "epoch": 0.6151626285095894, + "grad_norm": 1.9007022363034318, + "learning_rate": 1.3624406914251783e-06, + "loss": 1.0925, + "step": 5116 + }, + { + "epoch": 0.6152828714002284, + "grad_norm": 1.951319111413264, + "learning_rate": 1.3617024080005335e-06, + "loss": 1.1198, + "step": 5117 + }, + { + "epoch": 0.6154031142908676, + "grad_norm": 1.5645129398336277, + "learning_rate": 1.3609642214051266e-06, + "loss": 0.9812, + "step": 5118 + }, + { + "epoch": 0.6155233571815066, + "grad_norm": 2.243668655153575, + "learning_rate": 1.3602261317509385e-06, + "loss": 0.897, + "step": 5119 + }, + { + "epoch": 0.6156436000721457, + "grad_norm": 2.762968349887133, + "learning_rate": 1.3594881391499387e-06, + "loss": 1.0601, + "step": 5120 + }, + { + "epoch": 0.6157638429627849, + "grad_norm": 2.128572378313798, + "learning_rate": 1.3587502437140778e-06, + "loss": 1.0315, + "step": 5121 + }, + { + "epoch": 0.6158840858534239, + "grad_norm": 2.2631538331545147, + "learning_rate": 1.3580124455552952e-06, + "loss": 1.0872, + "step": 5122 + }, + { + "epoch": 0.616004328744063, + "grad_norm": 1.8865034087125272, + "learning_rate": 1.3572747447855148e-06, + "loss": 1.1127, + "step": 5123 + }, + { + "epoch": 0.6161245716347021, + "grad_norm": 2.0537080352080657, + "learning_rate": 1.356537141516644e-06, + "loss": 0.9274, + "step": 5124 + }, + { + "epoch": 0.6162448145253412, + "grad_norm": 2.1079873501001787, + "learning_rate": 1.3557996358605775e-06, + "loss": 0.8525, + "step": 5125 + }, + { + "epoch": 0.6163650574159802, + "grad_norm": 2.111397007216435, + "learning_rate": 1.3550622279291941e-06, + "loss": 0.9346, + "step": 5126 + }, + { + "epoch": 0.6164853003066194, + "grad_norm": 1.4472825928387087, + "learning_rate": 1.354324917834358e-06, + "loss": 1.0713, + "step": 5127 + }, + { + "epoch": 0.6166055431972585, + "grad_norm": 1.807949668276537, + "learning_rate": 1.353587705687918e-06, + "loss": 1.0049, + "step": 5128 + }, + { + "epoch": 0.6167257860878975, + "grad_norm": 2.8207151173134752, + "learning_rate": 1.3528505916017096e-06, + "loss": 0.9612, + "step": 5129 + }, + { + "epoch": 0.6168460289785367, + "grad_norm": 2.214894558224552, + "learning_rate": 1.3521135756875514e-06, + "loss": 1.1234, + "step": 5130 + }, + { + "epoch": 0.6169662718691757, + "grad_norm": 1.4704971473196637, + "learning_rate": 1.3513766580572496e-06, + "loss": 1.1003, + "step": 5131 + }, + { + "epoch": 0.6170865147598148, + "grad_norm": 2.1309328844518616, + "learning_rate": 1.3506398388225924e-06, + "loss": 1.0135, + "step": 5132 + }, + { + "epoch": 0.617206757650454, + "grad_norm": 2.0218438312657443, + "learning_rate": 1.349903118095355e-06, + "loss": 0.9558, + "step": 5133 + }, + { + "epoch": 0.617327000541093, + "grad_norm": 1.8866751529561157, + "learning_rate": 1.349166495987298e-06, + "loss": 0.9697, + "step": 5134 + }, + { + "epoch": 0.6174472434317321, + "grad_norm": 0.95266036918842, + "learning_rate": 1.348429972610166e-06, + "loss": 0.8467, + "step": 5135 + }, + { + "epoch": 0.6175674863223712, + "grad_norm": 0.8553412007041633, + "learning_rate": 1.3476935480756897e-06, + "loss": 0.8152, + "step": 5136 + }, + { + "epoch": 0.6176877292130103, + "grad_norm": 2.408215104295276, + "learning_rate": 1.346957222495583e-06, + "loss": 0.9903, + "step": 5137 + }, + { + "epoch": 0.6178079721036493, + "grad_norm": 2.3692706419136225, + "learning_rate": 1.3462209959815466e-06, + "loss": 0.9433, + "step": 5138 + }, + { + "epoch": 0.6179282149942885, + "grad_norm": 2.0089699344312573, + "learning_rate": 1.345484868645265e-06, + "loss": 0.9767, + "step": 5139 + }, + { + "epoch": 0.6180484578849276, + "grad_norm": 1.9091221168989336, + "learning_rate": 1.3447488405984088e-06, + "loss": 1.0213, + "step": 5140 + }, + { + "epoch": 0.6181687007755666, + "grad_norm": 3.7631738053042705, + "learning_rate": 1.3440129119526322e-06, + "loss": 0.9374, + "step": 5141 + }, + { + "epoch": 0.6182889436662057, + "grad_norm": 0.8062920729430925, + "learning_rate": 1.3432770828195762e-06, + "loss": 0.7493, + "step": 5142 + }, + { + "epoch": 0.6184091865568448, + "grad_norm": 2.396894822417931, + "learning_rate": 1.3425413533108635e-06, + "loss": 0.9431, + "step": 5143 + }, + { + "epoch": 0.6185294294474839, + "grad_norm": 2.3744369642542202, + "learning_rate": 1.341805723538105e-06, + "loss": 0.9422, + "step": 5144 + }, + { + "epoch": 0.618649672338123, + "grad_norm": 1.5366689591518246, + "learning_rate": 1.3410701936128948e-06, + "loss": 1.0113, + "step": 5145 + }, + { + "epoch": 0.6187699152287621, + "grad_norm": 2.7493874994831944, + "learning_rate": 1.340334763646812e-06, + "loss": 1.089, + "step": 5146 + }, + { + "epoch": 0.6188901581194012, + "grad_norm": 1.907036540873063, + "learning_rate": 1.3395994337514218e-06, + "loss": 0.981, + "step": 5147 + }, + { + "epoch": 0.6190104010100402, + "grad_norm": 2.698017565240937, + "learning_rate": 1.3388642040382725e-06, + "loss": 1.0221, + "step": 5148 + }, + { + "epoch": 0.6191306439006794, + "grad_norm": 1.7566648000216234, + "learning_rate": 1.3381290746188975e-06, + "loss": 1.0771, + "step": 5149 + }, + { + "epoch": 0.6192508867913185, + "grad_norm": 1.7473954846476072, + "learning_rate": 1.3373940456048152e-06, + "loss": 0.911, + "step": 5150 + }, + { + "epoch": 0.6193711296819575, + "grad_norm": 1.593870345415139, + "learning_rate": 1.3366591171075299e-06, + "loss": 0.8297, + "step": 5151 + }, + { + "epoch": 0.6194913725725967, + "grad_norm": 2.1509173144094857, + "learning_rate": 1.335924289238529e-06, + "loss": 1.1495, + "step": 5152 + }, + { + "epoch": 0.6196116154632357, + "grad_norm": 1.6904562733728634, + "learning_rate": 1.3351895621092859e-06, + "loss": 1.0065, + "step": 5153 + }, + { + "epoch": 0.6197318583538748, + "grad_norm": 2.1965110466477635, + "learning_rate": 1.3344549358312567e-06, + "loss": 0.9994, + "step": 5154 + }, + { + "epoch": 0.619852101244514, + "grad_norm": 2.5139066146403626, + "learning_rate": 1.3337204105158852e-06, + "loss": 1.021, + "step": 5155 + }, + { + "epoch": 0.619972344135153, + "grad_norm": 1.9463598656683636, + "learning_rate": 1.332985986274597e-06, + "loss": 0.9658, + "step": 5156 + }, + { + "epoch": 0.6200925870257921, + "grad_norm": 2.2313393198295244, + "learning_rate": 1.3322516632188047e-06, + "loss": 0.9875, + "step": 5157 + }, + { + "epoch": 0.6202128299164312, + "grad_norm": 2.293995408071031, + "learning_rate": 1.3315174414599045e-06, + "loss": 0.9027, + "step": 5158 + }, + { + "epoch": 0.6203330728070703, + "grad_norm": 1.8844166486954887, + "learning_rate": 1.3307833211092768e-06, + "loss": 0.9901, + "step": 5159 + }, + { + "epoch": 0.6204533156977093, + "grad_norm": 1.727932899655797, + "learning_rate": 1.3300493022782873e-06, + "loss": 0.9884, + "step": 5160 + }, + { + "epoch": 0.6205735585883485, + "grad_norm": 1.909144970512354, + "learning_rate": 1.3293153850782855e-06, + "loss": 0.9653, + "step": 5161 + }, + { + "epoch": 0.6206938014789876, + "grad_norm": 1.8451113443452718, + "learning_rate": 1.3285815696206069e-06, + "loss": 0.9495, + "step": 5162 + }, + { + "epoch": 0.6208140443696266, + "grad_norm": 2.9784949730329116, + "learning_rate": 1.32784785601657e-06, + "loss": 1.0062, + "step": 5163 + }, + { + "epoch": 0.6209342872602658, + "grad_norm": 1.798865501189466, + "learning_rate": 1.3271142443774798e-06, + "loss": 0.9789, + "step": 5164 + }, + { + "epoch": 0.6210545301509048, + "grad_norm": 1.865242875164696, + "learning_rate": 1.3263807348146228e-06, + "loss": 1.0525, + "step": 5165 + }, + { + "epoch": 0.6211747730415439, + "grad_norm": 2.1332543442924536, + "learning_rate": 1.3256473274392733e-06, + "loss": 0.9751, + "step": 5166 + }, + { + "epoch": 0.6212950159321831, + "grad_norm": 2.055210061579302, + "learning_rate": 1.3249140223626873e-06, + "loss": 0.937, + "step": 5167 + }, + { + "epoch": 0.6214152588228221, + "grad_norm": 1.9503788066607561, + "learning_rate": 1.3241808196961077e-06, + "loss": 0.9912, + "step": 5168 + }, + { + "epoch": 0.6215355017134612, + "grad_norm": 1.868907694717562, + "learning_rate": 1.3234477195507608e-06, + "loss": 0.9423, + "step": 5169 + }, + { + "epoch": 0.6216557446041003, + "grad_norm": 2.3940765795525856, + "learning_rate": 1.322714722037857e-06, + "loss": 0.8647, + "step": 5170 + }, + { + "epoch": 0.6217759874947394, + "grad_norm": 2.5976479895897797, + "learning_rate": 1.321981827268591e-06, + "loss": 1.0084, + "step": 5171 + }, + { + "epoch": 0.6218962303853784, + "grad_norm": 1.8939057566227713, + "learning_rate": 1.3212490353541426e-06, + "loss": 1.0493, + "step": 5172 + }, + { + "epoch": 0.6220164732760175, + "grad_norm": 2.0443508513192628, + "learning_rate": 1.3205163464056762e-06, + "loss": 1.043, + "step": 5173 + }, + { + "epoch": 0.6221367161666567, + "grad_norm": 1.7781528517648593, + "learning_rate": 1.319783760534339e-06, + "loss": 0.964, + "step": 5174 + }, + { + "epoch": 0.6222569590572957, + "grad_norm": 2.1090336276141324, + "learning_rate": 1.319051277851266e-06, + "loss": 0.9911, + "step": 5175 + }, + { + "epoch": 0.6223772019479348, + "grad_norm": 5.972420567415027, + "learning_rate": 1.3183188984675716e-06, + "loss": 1.0789, + "step": 5176 + }, + { + "epoch": 0.6224974448385739, + "grad_norm": 2.9187907704154474, + "learning_rate": 1.3175866224943586e-06, + "loss": 0.95, + "step": 5177 + }, + { + "epoch": 0.622617687729213, + "grad_norm": 2.642710567519031, + "learning_rate": 1.316854450042712e-06, + "loss": 0.9726, + "step": 5178 + }, + { + "epoch": 0.622737930619852, + "grad_norm": 1.9401709082130296, + "learning_rate": 1.3161223812237024e-06, + "loss": 0.9825, + "step": 5179 + }, + { + "epoch": 0.6228581735104912, + "grad_norm": 2.422328389706004, + "learning_rate": 1.3153904161483842e-06, + "loss": 1.0886, + "step": 5180 + }, + { + "epoch": 0.6229784164011303, + "grad_norm": 2.540149256608676, + "learning_rate": 1.3146585549277953e-06, + "loss": 1.0927, + "step": 5181 + }, + { + "epoch": 0.6230986592917693, + "grad_norm": 2.035741798673869, + "learning_rate": 1.3139267976729591e-06, + "loss": 1.0235, + "step": 5182 + }, + { + "epoch": 0.6232189021824085, + "grad_norm": 1.8441643976587447, + "learning_rate": 1.3131951444948815e-06, + "loss": 0.9535, + "step": 5183 + }, + { + "epoch": 0.6233391450730476, + "grad_norm": 2.4287525598313238, + "learning_rate": 1.3124635955045546e-06, + "loss": 0.9979, + "step": 5184 + }, + { + "epoch": 0.6234593879636866, + "grad_norm": 1.9051148834029992, + "learning_rate": 1.3117321508129537e-06, + "loss": 1.0767, + "step": 5185 + }, + { + "epoch": 0.6235796308543258, + "grad_norm": 1.6582481279320906, + "learning_rate": 1.3110008105310388e-06, + "loss": 0.9988, + "step": 5186 + }, + { + "epoch": 0.6236998737449648, + "grad_norm": 2.0147607414242334, + "learning_rate": 1.3102695747697526e-06, + "loss": 1.0183, + "step": 5187 + }, + { + "epoch": 0.6238201166356039, + "grad_norm": 3.390624578097972, + "learning_rate": 1.3095384436400237e-06, + "loss": 1.1404, + "step": 5188 + }, + { + "epoch": 0.623940359526243, + "grad_norm": 2.087599874151679, + "learning_rate": 1.3088074172527633e-06, + "loss": 1.0592, + "step": 5189 + }, + { + "epoch": 0.6240606024168821, + "grad_norm": 2.0648450524524127, + "learning_rate": 1.3080764957188684e-06, + "loss": 0.9524, + "step": 5190 + }, + { + "epoch": 0.6241808453075212, + "grad_norm": 1.7914725649471308, + "learning_rate": 1.3073456791492192e-06, + "loss": 0.9423, + "step": 5191 + }, + { + "epoch": 0.6243010881981603, + "grad_norm": 1.8619793516216765, + "learning_rate": 1.3066149676546801e-06, + "loss": 1.02, + "step": 5192 + }, + { + "epoch": 0.6244213310887994, + "grad_norm": 2.0696066750836453, + "learning_rate": 1.3058843613460985e-06, + "loss": 0.9009, + "step": 5193 + }, + { + "epoch": 0.6245415739794384, + "grad_norm": 1.9923476210927464, + "learning_rate": 1.3051538603343075e-06, + "loss": 0.9805, + "step": 5194 + }, + { + "epoch": 0.6246618168700776, + "grad_norm": 1.885140779940021, + "learning_rate": 1.3044234647301235e-06, + "loss": 0.9151, + "step": 5195 + }, + { + "epoch": 0.6247820597607167, + "grad_norm": 1.9574088762103312, + "learning_rate": 1.303693174644347e-06, + "loss": 0.9612, + "step": 5196 + }, + { + "epoch": 0.6249023026513557, + "grad_norm": 1.9315873033590916, + "learning_rate": 1.3029629901877625e-06, + "loss": 1.0443, + "step": 5197 + }, + { + "epoch": 0.6250225455419949, + "grad_norm": 2.6231890289803164, + "learning_rate": 1.3022329114711376e-06, + "loss": 1.0091, + "step": 5198 + }, + { + "epoch": 0.6251427884326339, + "grad_norm": 3.014264684434595, + "learning_rate": 1.3015029386052256e-06, + "loss": 0.934, + "step": 5199 + }, + { + "epoch": 0.625263031323273, + "grad_norm": 2.139265868033256, + "learning_rate": 1.3007730717007622e-06, + "loss": 0.9674, + "step": 5200 + }, + { + "epoch": 0.6253832742139122, + "grad_norm": 1.7755903041501828, + "learning_rate": 1.3000433108684676e-06, + "loss": 0.9932, + "step": 5201 + }, + { + "epoch": 0.6255035171045512, + "grad_norm": 2.365140426370112, + "learning_rate": 1.2993136562190467e-06, + "loss": 1.0347, + "step": 5202 + }, + { + "epoch": 0.6256237599951903, + "grad_norm": 1.8426252911021883, + "learning_rate": 1.2985841078631871e-06, + "loss": 0.9417, + "step": 5203 + }, + { + "epoch": 0.6257440028858293, + "grad_norm": 1.8801231329749934, + "learning_rate": 1.2978546659115608e-06, + "loss": 1.0217, + "step": 5204 + }, + { + "epoch": 0.6258642457764685, + "grad_norm": 2.060228194773233, + "learning_rate": 1.2971253304748228e-06, + "loss": 1.0895, + "step": 5205 + }, + { + "epoch": 0.6259844886671075, + "grad_norm": 1.5988184798355387, + "learning_rate": 1.296396101663614e-06, + "loss": 0.9853, + "step": 5206 + }, + { + "epoch": 0.6261047315577466, + "grad_norm": 2.1138483157492285, + "learning_rate": 1.2956669795885565e-06, + "loss": 1.0787, + "step": 5207 + }, + { + "epoch": 0.6262249744483858, + "grad_norm": 1.8934615743327698, + "learning_rate": 1.294937964360259e-06, + "loss": 0.9152, + "step": 5208 + }, + { + "epoch": 0.6263452173390248, + "grad_norm": 2.8040480030755974, + "learning_rate": 1.2942090560893108e-06, + "loss": 0.9493, + "step": 5209 + }, + { + "epoch": 0.6264654602296639, + "grad_norm": 6.026475034103263, + "learning_rate": 1.2934802548862882e-06, + "loss": 0.8457, + "step": 5210 + }, + { + "epoch": 0.626585703120303, + "grad_norm": 2.337048116649748, + "learning_rate": 1.292751560861749e-06, + "loss": 1.0642, + "step": 5211 + }, + { + "epoch": 0.6267059460109421, + "grad_norm": 1.795481929636068, + "learning_rate": 1.2920229741262354e-06, + "loss": 1.0326, + "step": 5212 + }, + { + "epoch": 0.6268261889015811, + "grad_norm": 2.1491611441099643, + "learning_rate": 1.2912944947902739e-06, + "loss": 0.9926, + "step": 5213 + }, + { + "epoch": 0.6269464317922203, + "grad_norm": 2.3033400912629656, + "learning_rate": 1.2905661229643742e-06, + "loss": 0.9541, + "step": 5214 + }, + { + "epoch": 0.6270666746828594, + "grad_norm": 2.6805691617138327, + "learning_rate": 1.2898378587590299e-06, + "loss": 1.0794, + "step": 5215 + }, + { + "epoch": 0.6271869175734984, + "grad_norm": 1.8257088630113036, + "learning_rate": 1.2891097022847173e-06, + "loss": 1.111, + "step": 5216 + }, + { + "epoch": 0.6273071604641376, + "grad_norm": 2.9233068537917433, + "learning_rate": 1.2883816536518978e-06, + "loss": 0.9045, + "step": 5217 + }, + { + "epoch": 0.6274274033547766, + "grad_norm": 2.1418691764509714, + "learning_rate": 1.2876537129710155e-06, + "loss": 1.0592, + "step": 5218 + }, + { + "epoch": 0.6275476462454157, + "grad_norm": 2.3076851697958083, + "learning_rate": 1.286925880352499e-06, + "loss": 0.9883, + "step": 5219 + }, + { + "epoch": 0.6276678891360549, + "grad_norm": 1.9506520599375479, + "learning_rate": 1.2861981559067592e-06, + "loss": 0.9505, + "step": 5220 + }, + { + "epoch": 0.6277881320266939, + "grad_norm": 1.883444272490079, + "learning_rate": 1.2854705397441917e-06, + "loss": 1.0381, + "step": 5221 + }, + { + "epoch": 0.627908374917333, + "grad_norm": 2.5214450398752155, + "learning_rate": 1.2847430319751747e-06, + "loss": 1.0131, + "step": 5222 + }, + { + "epoch": 0.6280286178079721, + "grad_norm": 2.2483406305819162, + "learning_rate": 1.2840156327100712e-06, + "loss": 0.9107, + "step": 5223 + }, + { + "epoch": 0.6281488606986112, + "grad_norm": 1.9165360019207376, + "learning_rate": 1.2832883420592272e-06, + "loss": 0.9634, + "step": 5224 + }, + { + "epoch": 0.6282691035892503, + "grad_norm": 2.4007412163238975, + "learning_rate": 1.282561160132972e-06, + "loss": 0.8777, + "step": 5225 + }, + { + "epoch": 0.6283893464798894, + "grad_norm": 2.28987247861595, + "learning_rate": 1.2818340870416186e-06, + "loss": 1.0456, + "step": 5226 + }, + { + "epoch": 0.6285095893705285, + "grad_norm": 1.9271289991861973, + "learning_rate": 1.2811071228954626e-06, + "loss": 0.9972, + "step": 5227 + }, + { + "epoch": 0.6286298322611675, + "grad_norm": 1.9140318576150535, + "learning_rate": 1.2803802678047846e-06, + "loss": 1.0473, + "step": 5228 + }, + { + "epoch": 0.6287500751518067, + "grad_norm": 2.677575883689394, + "learning_rate": 1.279653521879848e-06, + "loss": 0.9744, + "step": 5229 + }, + { + "epoch": 0.6288703180424458, + "grad_norm": 2.0586183028955616, + "learning_rate": 1.2789268852308997e-06, + "loss": 1.0753, + "step": 5230 + }, + { + "epoch": 0.6289905609330848, + "grad_norm": 2.0850619075009673, + "learning_rate": 1.2782003579681688e-06, + "loss": 0.9426, + "step": 5231 + }, + { + "epoch": 0.629110803823724, + "grad_norm": 1.6179876881756965, + "learning_rate": 1.2774739402018701e-06, + "loss": 0.9827, + "step": 5232 + }, + { + "epoch": 0.629231046714363, + "grad_norm": 2.87314213971479, + "learning_rate": 1.2767476320422002e-06, + "loss": 0.9696, + "step": 5233 + }, + { + "epoch": 0.6293512896050021, + "grad_norm": 0.6894110431742276, + "learning_rate": 1.2760214335993392e-06, + "loss": 0.8099, + "step": 5234 + }, + { + "epoch": 0.6294715324956413, + "grad_norm": 2.036374362733074, + "learning_rate": 1.2752953449834514e-06, + "loss": 0.8175, + "step": 5235 + }, + { + "epoch": 0.6295917753862803, + "grad_norm": 1.7916842910727753, + "learning_rate": 1.2745693663046836e-06, + "loss": 1.04, + "step": 5236 + }, + { + "epoch": 0.6297120182769194, + "grad_norm": 1.8307368864834532, + "learning_rate": 1.2738434976731662e-06, + "loss": 1.047, + "step": 5237 + }, + { + "epoch": 0.6298322611675584, + "grad_norm": 1.5928942308867329, + "learning_rate": 1.2731177391990125e-06, + "loss": 0.9873, + "step": 5238 + }, + { + "epoch": 0.6299525040581976, + "grad_norm": 2.556681656708555, + "learning_rate": 1.2723920909923203e-06, + "loss": 1.0531, + "step": 5239 + }, + { + "epoch": 0.6300727469488366, + "grad_norm": 0.8619904685456008, + "learning_rate": 1.2716665531631688e-06, + "loss": 0.8413, + "step": 5240 + }, + { + "epoch": 0.6301929898394757, + "grad_norm": 2.7467285117189584, + "learning_rate": 1.270941125821623e-06, + "loss": 1.0119, + "step": 5241 + }, + { + "epoch": 0.6303132327301149, + "grad_norm": 1.6962589136056718, + "learning_rate": 1.2702158090777278e-06, + "loss": 0.9868, + "step": 5242 + }, + { + "epoch": 0.6304334756207539, + "grad_norm": 2.117343010065145, + "learning_rate": 1.2694906030415148e-06, + "loss": 0.9872, + "step": 5243 + }, + { + "epoch": 0.630553718511393, + "grad_norm": 2.723046952049113, + "learning_rate": 1.2687655078229958e-06, + "loss": 1.0547, + "step": 5244 + }, + { + "epoch": 0.6306739614020321, + "grad_norm": 2.3138623735125496, + "learning_rate": 1.2680405235321678e-06, + "loss": 0.9309, + "step": 5245 + }, + { + "epoch": 0.6307942042926712, + "grad_norm": 2.075538574551358, + "learning_rate": 1.267315650279011e-06, + "loss": 1.0244, + "step": 5246 + }, + { + "epoch": 0.6309144471833102, + "grad_norm": 1.9233302987513852, + "learning_rate": 1.2665908881734874e-06, + "loss": 0.9773, + "step": 5247 + }, + { + "epoch": 0.6310346900739494, + "grad_norm": 2.143897902833397, + "learning_rate": 1.2658662373255432e-06, + "loss": 1.0859, + "step": 5248 + }, + { + "epoch": 0.6311549329645885, + "grad_norm": 0.7140736418341033, + "learning_rate": 1.2651416978451063e-06, + "loss": 0.7606, + "step": 5249 + }, + { + "epoch": 0.6312751758552275, + "grad_norm": 1.9806877180710072, + "learning_rate": 1.2644172698420903e-06, + "loss": 0.8885, + "step": 5250 + }, + { + "epoch": 0.6313954187458667, + "grad_norm": 1.7981572095991356, + "learning_rate": 1.2636929534263892e-06, + "loss": 1.0867, + "step": 5251 + }, + { + "epoch": 0.6315156616365057, + "grad_norm": 2.2958780837039092, + "learning_rate": 1.2629687487078821e-06, + "loss": 1.0065, + "step": 5252 + }, + { + "epoch": 0.6316359045271448, + "grad_norm": 2.049585658743756, + "learning_rate": 1.2622446557964293e-06, + "loss": 0.9982, + "step": 5253 + }, + { + "epoch": 0.631756147417784, + "grad_norm": 1.8433672620859627, + "learning_rate": 1.261520674801876e-06, + "loss": 0.9503, + "step": 5254 + }, + { + "epoch": 0.631876390308423, + "grad_norm": 2.1457004938086905, + "learning_rate": 1.2607968058340488e-06, + "loss": 0.9578, + "step": 5255 + }, + { + "epoch": 0.6319966331990621, + "grad_norm": 2.8976868842535084, + "learning_rate": 1.2600730490027583e-06, + "loss": 0.9688, + "step": 5256 + }, + { + "epoch": 0.6321168760897012, + "grad_norm": 1.6041656923497531, + "learning_rate": 1.2593494044177984e-06, + "loss": 1.0404, + "step": 5257 + }, + { + "epoch": 0.6322371189803403, + "grad_norm": 2.6319216344466194, + "learning_rate": 1.2586258721889448e-06, + "loss": 1.0458, + "step": 5258 + }, + { + "epoch": 0.6323573618709794, + "grad_norm": 2.0884328482446346, + "learning_rate": 1.2579024524259573e-06, + "loss": 1.0529, + "step": 5259 + }, + { + "epoch": 0.6324776047616185, + "grad_norm": 2.19796321391735, + "learning_rate": 1.2571791452385768e-06, + "loss": 1.1506, + "step": 5260 + }, + { + "epoch": 0.6325978476522576, + "grad_norm": 1.7012342404344019, + "learning_rate": 1.2564559507365301e-06, + "loss": 1.0105, + "step": 5261 + }, + { + "epoch": 0.6327180905428966, + "grad_norm": 2.303463782312051, + "learning_rate": 1.2557328690295244e-06, + "loss": 1.0257, + "step": 5262 + }, + { + "epoch": 0.6328383334335358, + "grad_norm": 1.7538083055501184, + "learning_rate": 1.255009900227251e-06, + "loss": 1.0016, + "step": 5263 + }, + { + "epoch": 0.6329585763241748, + "grad_norm": 1.7125105920171695, + "learning_rate": 1.254287044439383e-06, + "loss": 1.0314, + "step": 5264 + }, + { + "epoch": 0.6330788192148139, + "grad_norm": 0.7988539041029689, + "learning_rate": 1.2535643017755776e-06, + "loss": 0.7818, + "step": 5265 + }, + { + "epoch": 0.6331990621054531, + "grad_norm": 2.2950438381820173, + "learning_rate": 1.2528416723454737e-06, + "loss": 0.9573, + "step": 5266 + }, + { + "epoch": 0.6333193049960921, + "grad_norm": 2.1863850613245273, + "learning_rate": 1.2521191562586945e-06, + "loss": 0.9489, + "step": 5267 + }, + { + "epoch": 0.6334395478867312, + "grad_norm": 10.738000249256253, + "learning_rate": 1.2513967536248445e-06, + "loss": 1.0055, + "step": 5268 + }, + { + "epoch": 0.6335597907773702, + "grad_norm": 1.7088648814251346, + "learning_rate": 1.2506744645535117e-06, + "loss": 1.0477, + "step": 5269 + }, + { + "epoch": 0.6336800336680094, + "grad_norm": 1.9377965392556746, + "learning_rate": 1.249952289154267e-06, + "loss": 0.8389, + "step": 5270 + }, + { + "epoch": 0.6338002765586485, + "grad_norm": 1.9725117545508901, + "learning_rate": 1.2492302275366635e-06, + "loss": 1.0032, + "step": 5271 + }, + { + "epoch": 0.6339205194492875, + "grad_norm": 2.1080880088964915, + "learning_rate": 1.2485082798102377e-06, + "loss": 0.8901, + "step": 5272 + }, + { + "epoch": 0.6340407623399267, + "grad_norm": 2.21194208729487, + "learning_rate": 1.2477864460845084e-06, + "loss": 0.9197, + "step": 5273 + }, + { + "epoch": 0.6341610052305657, + "grad_norm": 2.8629278404637004, + "learning_rate": 1.2470647264689776e-06, + "loss": 0.9632, + "step": 5274 + }, + { + "epoch": 0.6342812481212048, + "grad_norm": 2.2744499128373565, + "learning_rate": 1.2463431210731282e-06, + "loss": 0.9515, + "step": 5275 + }, + { + "epoch": 0.634401491011844, + "grad_norm": 2.3188227228240743, + "learning_rate": 1.2456216300064289e-06, + "loss": 0.9982, + "step": 5276 + }, + { + "epoch": 0.634521733902483, + "grad_norm": 3.2217644954641544, + "learning_rate": 1.244900253378328e-06, + "loss": 1.0229, + "step": 5277 + }, + { + "epoch": 0.6346419767931221, + "grad_norm": 1.9823095430766493, + "learning_rate": 1.2441789912982583e-06, + "loss": 0.9269, + "step": 5278 + }, + { + "epoch": 0.6347622196837612, + "grad_norm": 1.944041744796514, + "learning_rate": 1.2434578438756346e-06, + "loss": 0.8856, + "step": 5279 + }, + { + "epoch": 0.6348824625744003, + "grad_norm": 1.9108230269573219, + "learning_rate": 1.242736811219855e-06, + "loss": 1.0158, + "step": 5280 + }, + { + "epoch": 0.6350027054650393, + "grad_norm": 2.2963170003071567, + "learning_rate": 1.2420158934402988e-06, + "loss": 1.0559, + "step": 5281 + }, + { + "epoch": 0.6351229483556785, + "grad_norm": 1.8859977529235268, + "learning_rate": 1.2412950906463286e-06, + "loss": 1.0858, + "step": 5282 + }, + { + "epoch": 0.6352431912463176, + "grad_norm": 2.066871749280156, + "learning_rate": 1.2405744029472902e-06, + "loss": 1.1355, + "step": 5283 + }, + { + "epoch": 0.6353634341369566, + "grad_norm": 1.983174719465238, + "learning_rate": 1.2398538304525108e-06, + "loss": 0.9961, + "step": 5284 + }, + { + "epoch": 0.6354836770275958, + "grad_norm": 2.262699738005491, + "learning_rate": 1.2391333732713016e-06, + "loss": 0.9916, + "step": 5285 + }, + { + "epoch": 0.6356039199182348, + "grad_norm": 2.1099559972794033, + "learning_rate": 1.2384130315129543e-06, + "loss": 1.021, + "step": 5286 + }, + { + "epoch": 0.6357241628088739, + "grad_norm": 3.151587337609737, + "learning_rate": 1.2376928052867447e-06, + "loss": 0.9738, + "step": 5287 + }, + { + "epoch": 0.6358444056995131, + "grad_norm": 3.058657200960225, + "learning_rate": 1.2369726947019299e-06, + "loss": 1.0103, + "step": 5288 + }, + { + "epoch": 0.6359646485901521, + "grad_norm": 2.3703857571491, + "learning_rate": 1.2362526998677511e-06, + "loss": 0.9102, + "step": 5289 + }, + { + "epoch": 0.6360848914807912, + "grad_norm": 1.976347900320641, + "learning_rate": 1.2355328208934301e-06, + "loss": 1.0794, + "step": 5290 + }, + { + "epoch": 0.6362051343714303, + "grad_norm": 1.892232920497643, + "learning_rate": 1.2348130578881728e-06, + "loss": 0.9585, + "step": 5291 + }, + { + "epoch": 0.6363253772620694, + "grad_norm": 3.0284914659087336, + "learning_rate": 1.2340934109611664e-06, + "loss": 1.0019, + "step": 5292 + }, + { + "epoch": 0.6364456201527084, + "grad_norm": 3.2967533613427777, + "learning_rate": 1.2333738802215798e-06, + "loss": 0.9234, + "step": 5293 + }, + { + "epoch": 0.6365658630433476, + "grad_norm": 1.8408967262820415, + "learning_rate": 1.2326544657785668e-06, + "loss": 1.045, + "step": 5294 + }, + { + "epoch": 0.6366861059339867, + "grad_norm": 2.2407310088337185, + "learning_rate": 1.2319351677412608e-06, + "loss": 0.9795, + "step": 5295 + }, + { + "epoch": 0.6368063488246257, + "grad_norm": 2.794154287694429, + "learning_rate": 1.2312159862187796e-06, + "loss": 0.9805, + "step": 5296 + }, + { + "epoch": 0.6369265917152649, + "grad_norm": 1.5960238821876782, + "learning_rate": 1.2304969213202217e-06, + "loss": 0.9997, + "step": 5297 + }, + { + "epoch": 0.6370468346059039, + "grad_norm": 3.5633125047874907, + "learning_rate": 1.2297779731546692e-06, + "loss": 1.024, + "step": 5298 + }, + { + "epoch": 0.637167077496543, + "grad_norm": 2.230951634394463, + "learning_rate": 1.2290591418311853e-06, + "loss": 1.0186, + "step": 5299 + }, + { + "epoch": 0.637287320387182, + "grad_norm": 1.6581507877856525, + "learning_rate": 1.2283404274588172e-06, + "loss": 0.9576, + "step": 5300 + }, + { + "epoch": 0.6374075632778212, + "grad_norm": 0.749899579318893, + "learning_rate": 1.227621830146592e-06, + "loss": 0.7649, + "step": 5301 + }, + { + "epoch": 0.6375278061684603, + "grad_norm": 1.8659011048648817, + "learning_rate": 1.2269033500035217e-06, + "loss": 1.0284, + "step": 5302 + }, + { + "epoch": 0.6376480490590993, + "grad_norm": 1.8674230067915933, + "learning_rate": 1.2261849871385988e-06, + "loss": 0.9767, + "step": 5303 + }, + { + "epoch": 0.6377682919497385, + "grad_norm": 2.0203922878479945, + "learning_rate": 1.2254667416607972e-06, + "loss": 0.8614, + "step": 5304 + }, + { + "epoch": 0.6378885348403776, + "grad_norm": 1.8691612250489362, + "learning_rate": 1.2247486136790756e-06, + "loss": 1.0687, + "step": 5305 + }, + { + "epoch": 0.6380087777310166, + "grad_norm": 2.7378437819237877, + "learning_rate": 1.2240306033023726e-06, + "loss": 1.0386, + "step": 5306 + }, + { + "epoch": 0.6381290206216558, + "grad_norm": 1.984176745986598, + "learning_rate": 1.223312710639611e-06, + "loss": 0.9555, + "step": 5307 + }, + { + "epoch": 0.6382492635122948, + "grad_norm": 2.1200785024971336, + "learning_rate": 1.2225949357996928e-06, + "loss": 1.1031, + "step": 5308 + }, + { + "epoch": 0.6383695064029339, + "grad_norm": 1.5647954482928972, + "learning_rate": 1.221877278891505e-06, + "loss": 1.0416, + "step": 5309 + }, + { + "epoch": 0.638489749293573, + "grad_norm": 2.0998756417283087, + "learning_rate": 1.221159740023915e-06, + "loss": 0.9529, + "step": 5310 + }, + { + "epoch": 0.6386099921842121, + "grad_norm": 4.586564237278875, + "learning_rate": 1.2204423193057735e-06, + "loss": 0.9662, + "step": 5311 + }, + { + "epoch": 0.6387302350748512, + "grad_norm": 0.8639881467646338, + "learning_rate": 1.2197250168459122e-06, + "loss": 0.8708, + "step": 5312 + }, + { + "epoch": 0.6388504779654903, + "grad_norm": 2.0830057522405423, + "learning_rate": 1.2190078327531454e-06, + "loss": 0.9824, + "step": 5313 + }, + { + "epoch": 0.6389707208561294, + "grad_norm": 1.6278344156800582, + "learning_rate": 1.2182907671362697e-06, + "loss": 0.9642, + "step": 5314 + }, + { + "epoch": 0.6390909637467684, + "grad_norm": 2.2176648562284567, + "learning_rate": 1.2175738201040626e-06, + "loss": 1.0226, + "step": 5315 + }, + { + "epoch": 0.6392112066374076, + "grad_norm": 1.949959959939545, + "learning_rate": 1.2168569917652855e-06, + "loss": 1.0235, + "step": 5316 + }, + { + "epoch": 0.6393314495280467, + "grad_norm": 1.5993698369278628, + "learning_rate": 1.2161402822286797e-06, + "loss": 0.8758, + "step": 5317 + }, + { + "epoch": 0.6394516924186857, + "grad_norm": 1.8897959254861703, + "learning_rate": 1.2154236916029703e-06, + "loss": 1.0247, + "step": 5318 + }, + { + "epoch": 0.6395719353093249, + "grad_norm": 2.403093521844592, + "learning_rate": 1.2147072199968627e-06, + "loss": 0.9683, + "step": 5319 + }, + { + "epoch": 0.6396921781999639, + "grad_norm": 1.8739401046886994, + "learning_rate": 1.2139908675190454e-06, + "loss": 0.9544, + "step": 5320 + }, + { + "epoch": 0.639812421090603, + "grad_norm": 2.1831856825700506, + "learning_rate": 1.2132746342781883e-06, + "loss": 0.9839, + "step": 5321 + }, + { + "epoch": 0.6399326639812422, + "grad_norm": 2.477000778358285, + "learning_rate": 1.2125585203829442e-06, + "loss": 1.0366, + "step": 5322 + }, + { + "epoch": 0.6400529068718812, + "grad_norm": 1.988524416460205, + "learning_rate": 1.211842525941946e-06, + "loss": 0.9789, + "step": 5323 + }, + { + "epoch": 0.6401731497625203, + "grad_norm": 2.1662475351714714, + "learning_rate": 1.2111266510638105e-06, + "loss": 1.0281, + "step": 5324 + }, + { + "epoch": 0.6402933926531594, + "grad_norm": 1.8565725797627741, + "learning_rate": 1.2104108958571346e-06, + "loss": 1.0383, + "step": 5325 + }, + { + "epoch": 0.6404136355437985, + "grad_norm": 1.8703266395028002, + "learning_rate": 1.2096952604304975e-06, + "loss": 0.9935, + "step": 5326 + }, + { + "epoch": 0.6405338784344375, + "grad_norm": 2.1998964285312725, + "learning_rate": 1.2089797448924616e-06, + "loss": 0.9404, + "step": 5327 + }, + { + "epoch": 0.6406541213250767, + "grad_norm": 2.0093563096762703, + "learning_rate": 1.2082643493515692e-06, + "loss": 0.8948, + "step": 5328 + }, + { + "epoch": 0.6407743642157158, + "grad_norm": 1.8475421765459674, + "learning_rate": 1.207549073916346e-06, + "loss": 1.0557, + "step": 5329 + }, + { + "epoch": 0.6408946071063548, + "grad_norm": 2.250633362529699, + "learning_rate": 1.2068339186952976e-06, + "loss": 1.0154, + "step": 5330 + }, + { + "epoch": 0.6410148499969939, + "grad_norm": 2.338795830070128, + "learning_rate": 1.2061188837969136e-06, + "loss": 0.9675, + "step": 5331 + }, + { + "epoch": 0.641135092887633, + "grad_norm": 2.4296181230360876, + "learning_rate": 1.2054039693296631e-06, + "loss": 1.0754, + "step": 5332 + }, + { + "epoch": 0.6412553357782721, + "grad_norm": 1.791495721631447, + "learning_rate": 1.2046891754019992e-06, + "loss": 1.0514, + "step": 5333 + }, + { + "epoch": 0.6413755786689112, + "grad_norm": 2.0983460498007367, + "learning_rate": 1.2039745021223548e-06, + "loss": 1.0631, + "step": 5334 + }, + { + "epoch": 0.6414958215595503, + "grad_norm": 0.8102619819473148, + "learning_rate": 1.2032599495991456e-06, + "loss": 0.8083, + "step": 5335 + }, + { + "epoch": 0.6416160644501894, + "grad_norm": 1.7332756063434245, + "learning_rate": 1.2025455179407685e-06, + "loss": 0.9365, + "step": 5336 + }, + { + "epoch": 0.6417363073408284, + "grad_norm": 2.009145331420441, + "learning_rate": 1.2018312072556022e-06, + "loss": 0.9742, + "step": 5337 + }, + { + "epoch": 0.6418565502314676, + "grad_norm": 1.8350281106734223, + "learning_rate": 1.2011170176520077e-06, + "loss": 0.9786, + "step": 5338 + }, + { + "epoch": 0.6419767931221066, + "grad_norm": 2.004360809228358, + "learning_rate": 1.2004029492383256e-06, + "loss": 1.0529, + "step": 5339 + }, + { + "epoch": 0.6420970360127457, + "grad_norm": 2.0949612430456277, + "learning_rate": 1.1996890021228814e-06, + "loss": 0.9729, + "step": 5340 + }, + { + "epoch": 0.6422172789033849, + "grad_norm": 1.5266257547257827, + "learning_rate": 1.1989751764139785e-06, + "loss": 0.937, + "step": 5341 + }, + { + "epoch": 0.6423375217940239, + "grad_norm": 1.6014539728377835, + "learning_rate": 1.1982614722199044e-06, + "loss": 1.065, + "step": 5342 + }, + { + "epoch": 0.642457764684663, + "grad_norm": 2.131600171628585, + "learning_rate": 1.1975478896489276e-06, + "loss": 1.0118, + "step": 5343 + }, + { + "epoch": 0.6425780075753021, + "grad_norm": 2.1566795253026396, + "learning_rate": 1.1968344288092981e-06, + "loss": 1.0021, + "step": 5344 + }, + { + "epoch": 0.6426982504659412, + "grad_norm": 1.7961334025024314, + "learning_rate": 1.1961210898092468e-06, + "loss": 0.8853, + "step": 5345 + }, + { + "epoch": 0.6428184933565803, + "grad_norm": 2.262482772677259, + "learning_rate": 1.1954078727569874e-06, + "loss": 1.0287, + "step": 5346 + }, + { + "epoch": 0.6429387362472194, + "grad_norm": 1.8261682194147904, + "learning_rate": 1.1946947777607141e-06, + "loss": 1.0177, + "step": 5347 + }, + { + "epoch": 0.6430589791378585, + "grad_norm": 2.2401253924014255, + "learning_rate": 1.1939818049286024e-06, + "loss": 1.0398, + "step": 5348 + }, + { + "epoch": 0.6431792220284975, + "grad_norm": 1.6814391472927541, + "learning_rate": 1.1932689543688101e-06, + "loss": 0.9921, + "step": 5349 + }, + { + "epoch": 0.6432994649191367, + "grad_norm": 1.8262961605434167, + "learning_rate": 1.1925562261894756e-06, + "loss": 0.9595, + "step": 5350 + }, + { + "epoch": 0.6434197078097758, + "grad_norm": 1.815112565913714, + "learning_rate": 1.1918436204987207e-06, + "loss": 1.0127, + "step": 5351 + }, + { + "epoch": 0.6435399507004148, + "grad_norm": 2.2188609995410764, + "learning_rate": 1.191131137404645e-06, + "loss": 1.0516, + "step": 5352 + }, + { + "epoch": 0.643660193591054, + "grad_norm": 2.672428363807003, + "learning_rate": 1.190418777015333e-06, + "loss": 1.0089, + "step": 5353 + }, + { + "epoch": 0.643780436481693, + "grad_norm": 1.59456097782128, + "learning_rate": 1.1897065394388487e-06, + "loss": 0.9753, + "step": 5354 + }, + { + "epoch": 0.6439006793723321, + "grad_norm": 1.6243133561406087, + "learning_rate": 1.1889944247832385e-06, + "loss": 1.0021, + "step": 5355 + }, + { + "epoch": 0.6440209222629713, + "grad_norm": 2.2885856261883664, + "learning_rate": 1.1882824331565283e-06, + "loss": 0.946, + "step": 5356 + }, + { + "epoch": 0.6441411651536103, + "grad_norm": 2.303904979210028, + "learning_rate": 1.1875705646667287e-06, + "loss": 1.129, + "step": 5357 + }, + { + "epoch": 0.6442614080442494, + "grad_norm": 2.6396236891297296, + "learning_rate": 1.1868588194218282e-06, + "loss": 0.9927, + "step": 5358 + }, + { + "epoch": 0.6443816509348885, + "grad_norm": 1.8125064455115054, + "learning_rate": 1.1861471975297979e-06, + "loss": 0.9775, + "step": 5359 + }, + { + "epoch": 0.6445018938255276, + "grad_norm": 50.26148908614079, + "learning_rate": 1.185435699098591e-06, + "loss": 0.9431, + "step": 5360 + }, + { + "epoch": 0.6446221367161666, + "grad_norm": 2.590913991604053, + "learning_rate": 1.1847243242361403e-06, + "loss": 1.02, + "step": 5361 + }, + { + "epoch": 0.6447423796068057, + "grad_norm": 1.7111752140263559, + "learning_rate": 1.1840130730503624e-06, + "loss": 1.0162, + "step": 5362 + }, + { + "epoch": 0.6448626224974449, + "grad_norm": 1.7533130619680624, + "learning_rate": 1.1833019456491518e-06, + "loss": 0.9843, + "step": 5363 + }, + { + "epoch": 0.6449828653880839, + "grad_norm": 2.24116593858012, + "learning_rate": 1.1825909421403871e-06, + "loss": 1.0231, + "step": 5364 + }, + { + "epoch": 0.645103108278723, + "grad_norm": 2.0860375083923293, + "learning_rate": 1.181880062631926e-06, + "loss": 0.998, + "step": 5365 + }, + { + "epoch": 0.6452233511693621, + "grad_norm": 3.1808565507267486, + "learning_rate": 1.1811693072316093e-06, + "loss": 1.0843, + "step": 5366 + }, + { + "epoch": 0.6453435940600012, + "grad_norm": 2.590466914858226, + "learning_rate": 1.1804586760472574e-06, + "loss": 1.0816, + "step": 5367 + }, + { + "epoch": 0.6454638369506402, + "grad_norm": 2.9336555159526307, + "learning_rate": 1.1797481691866736e-06, + "loss": 1.0367, + "step": 5368 + }, + { + "epoch": 0.6455840798412794, + "grad_norm": 2.27067424205219, + "learning_rate": 1.1790377867576393e-06, + "loss": 1.0675, + "step": 5369 + }, + { + "epoch": 0.6457043227319185, + "grad_norm": 1.8324101753650164, + "learning_rate": 1.1783275288679203e-06, + "loss": 1.005, + "step": 5370 + }, + { + "epoch": 0.6458245656225575, + "grad_norm": 0.8557065171726808, + "learning_rate": 1.177617395625262e-06, + "loss": 0.8089, + "step": 5371 + }, + { + "epoch": 0.6459448085131967, + "grad_norm": 1.8333825046996652, + "learning_rate": 1.1769073871373908e-06, + "loss": 0.992, + "step": 5372 + }, + { + "epoch": 0.6460650514038357, + "grad_norm": 1.8112636822064596, + "learning_rate": 1.176197503512015e-06, + "loss": 1.0778, + "step": 5373 + }, + { + "epoch": 0.6461852942944748, + "grad_norm": 2.559750826749343, + "learning_rate": 1.1754877448568223e-06, + "loss": 1.0593, + "step": 5374 + }, + { + "epoch": 0.646305537185114, + "grad_norm": 2.239101718897192, + "learning_rate": 1.1747781112794837e-06, + "loss": 1.1361, + "step": 5375 + }, + { + "epoch": 0.646425780075753, + "grad_norm": 1.7420611827542765, + "learning_rate": 1.1740686028876487e-06, + "loss": 1.0653, + "step": 5376 + }, + { + "epoch": 0.6465460229663921, + "grad_norm": 2.801566680207248, + "learning_rate": 1.1733592197889507e-06, + "loss": 0.9851, + "step": 5377 + }, + { + "epoch": 0.6466662658570312, + "grad_norm": 1.9413241516807036, + "learning_rate": 1.1726499620910014e-06, + "loss": 0.9644, + "step": 5378 + }, + { + "epoch": 0.6467865087476703, + "grad_norm": 2.015368659700546, + "learning_rate": 1.1719408299013955e-06, + "loss": 1.0174, + "step": 5379 + }, + { + "epoch": 0.6469067516383094, + "grad_norm": 2.4494926628007576, + "learning_rate": 1.1712318233277067e-06, + "loss": 0.9907, + "step": 5380 + }, + { + "epoch": 0.6470269945289485, + "grad_norm": 0.7707621223617428, + "learning_rate": 1.1705229424774916e-06, + "loss": 0.8157, + "step": 5381 + }, + { + "epoch": 0.6471472374195876, + "grad_norm": 1.7791463911820475, + "learning_rate": 1.1698141874582867e-06, + "loss": 0.8793, + "step": 5382 + }, + { + "epoch": 0.6472674803102266, + "grad_norm": 1.9090759093038001, + "learning_rate": 1.169105558377609e-06, + "loss": 0.9603, + "step": 5383 + }, + { + "epoch": 0.6473877232008658, + "grad_norm": 2.1471641512386754, + "learning_rate": 1.1683970553429587e-06, + "loss": 1.019, + "step": 5384 + }, + { + "epoch": 0.6475079660915048, + "grad_norm": 2.002488138304213, + "learning_rate": 1.1676886784618128e-06, + "loss": 1.0573, + "step": 5385 + }, + { + "epoch": 0.6476282089821439, + "grad_norm": 2.2436843819971797, + "learning_rate": 1.1669804278416332e-06, + "loss": 1.0734, + "step": 5386 + }, + { + "epoch": 0.6477484518727831, + "grad_norm": 3.798286081589007, + "learning_rate": 1.1662723035898602e-06, + "loss": 0.9512, + "step": 5387 + }, + { + "epoch": 0.6478686947634221, + "grad_norm": 1.7552634512588687, + "learning_rate": 1.165564305813915e-06, + "loss": 1.0566, + "step": 5388 + }, + { + "epoch": 0.6479889376540612, + "grad_norm": 1.8472122400853845, + "learning_rate": 1.1648564346212019e-06, + "loss": 1.0481, + "step": 5389 + }, + { + "epoch": 0.6481091805447003, + "grad_norm": 1.7859280376567463, + "learning_rate": 1.164148690119104e-06, + "loss": 1.0004, + "step": 5390 + }, + { + "epoch": 0.6482294234353394, + "grad_norm": 1.7420254619152205, + "learning_rate": 1.163441072414985e-06, + "loss": 0.9753, + "step": 5391 + }, + { + "epoch": 0.6483496663259785, + "grad_norm": 2.135776235189571, + "learning_rate": 1.16273358161619e-06, + "loss": 0.9365, + "step": 5392 + }, + { + "epoch": 0.6484699092166175, + "grad_norm": 2.5247005927209947, + "learning_rate": 1.1620262178300446e-06, + "loss": 1.0707, + "step": 5393 + }, + { + "epoch": 0.6485901521072567, + "grad_norm": 1.8830331380260994, + "learning_rate": 1.1613189811638563e-06, + "loss": 0.9991, + "step": 5394 + }, + { + "epoch": 0.6487103949978957, + "grad_norm": 1.6637845232539386, + "learning_rate": 1.1606118717249117e-06, + "loss": 1.0175, + "step": 5395 + }, + { + "epoch": 0.6488306378885348, + "grad_norm": 2.040005403773313, + "learning_rate": 1.1599048896204787e-06, + "loss": 0.9114, + "step": 5396 + }, + { + "epoch": 0.648950880779174, + "grad_norm": 2.332152805962265, + "learning_rate": 1.1591980349578061e-06, + "loss": 1.0453, + "step": 5397 + }, + { + "epoch": 0.649071123669813, + "grad_norm": 0.7399476635957707, + "learning_rate": 1.158491307844123e-06, + "loss": 0.7808, + "step": 5398 + }, + { + "epoch": 0.6491913665604521, + "grad_norm": 1.745808896035603, + "learning_rate": 1.1577847083866387e-06, + "loss": 1.0778, + "step": 5399 + }, + { + "epoch": 0.6493116094510912, + "grad_norm": 1.851106096504892, + "learning_rate": 1.1570782366925453e-06, + "loss": 0.9588, + "step": 5400 + }, + { + "epoch": 0.6494318523417303, + "grad_norm": 22.060093043278965, + "learning_rate": 1.1563718928690132e-06, + "loss": 0.9961, + "step": 5401 + }, + { + "epoch": 0.6495520952323693, + "grad_norm": 2.0047936212960322, + "learning_rate": 1.1556656770231942e-06, + "loss": 0.954, + "step": 5402 + }, + { + "epoch": 0.6496723381230085, + "grad_norm": 1.639543013054703, + "learning_rate": 1.1549595892622207e-06, + "loss": 0.994, + "step": 5403 + }, + { + "epoch": 0.6497925810136476, + "grad_norm": 0.8313883056883574, + "learning_rate": 1.1542536296932047e-06, + "loss": 0.8278, + "step": 5404 + }, + { + "epoch": 0.6499128239042866, + "grad_norm": 2.5904272466448295, + "learning_rate": 1.1535477984232414e-06, + "loss": 0.9364, + "step": 5405 + }, + { + "epoch": 0.6500330667949258, + "grad_norm": 2.9271354444397026, + "learning_rate": 1.152842095559404e-06, + "loss": 1.007, + "step": 5406 + }, + { + "epoch": 0.6501533096855648, + "grad_norm": 1.7404205530933146, + "learning_rate": 1.1521365212087474e-06, + "loss": 1.0087, + "step": 5407 + }, + { + "epoch": 0.6502735525762039, + "grad_norm": 1.6418753083123134, + "learning_rate": 1.1514310754783062e-06, + "loss": 0.9451, + "step": 5408 + }, + { + "epoch": 0.6503937954668431, + "grad_norm": 1.9744466682509336, + "learning_rate": 1.1507257584750964e-06, + "loss": 0.9687, + "step": 5409 + }, + { + "epoch": 0.6505140383574821, + "grad_norm": 1.853875819622417, + "learning_rate": 1.150020570306113e-06, + "loss": 1.0074, + "step": 5410 + }, + { + "epoch": 0.6506342812481212, + "grad_norm": 3.0729683219749155, + "learning_rate": 1.1493155110783338e-06, + "loss": 0.988, + "step": 5411 + }, + { + "epoch": 0.6507545241387603, + "grad_norm": 2.078239150963481, + "learning_rate": 1.1486105808987155e-06, + "loss": 0.944, + "step": 5412 + }, + { + "epoch": 0.6508747670293994, + "grad_norm": 1.977035887952517, + "learning_rate": 1.1479057798741947e-06, + "loss": 1.05, + "step": 5413 + }, + { + "epoch": 0.6509950099200384, + "grad_norm": 0.790248470586101, + "learning_rate": 1.14720110811169e-06, + "loss": 0.7705, + "step": 5414 + }, + { + "epoch": 0.6511152528106776, + "grad_norm": 2.183791368307744, + "learning_rate": 1.146496565718098e-06, + "loss": 1.0047, + "step": 5415 + }, + { + "epoch": 0.6512354957013167, + "grad_norm": 2.9944355540083634, + "learning_rate": 1.1457921528002996e-06, + "loss": 0.9931, + "step": 5416 + }, + { + "epoch": 0.6513557385919557, + "grad_norm": 2.9643501202518254, + "learning_rate": 1.1450878694651522e-06, + "loss": 0.956, + "step": 5417 + }, + { + "epoch": 0.6514759814825949, + "grad_norm": 2.6782268847533848, + "learning_rate": 1.1443837158194954e-06, + "loss": 0.869, + "step": 5418 + }, + { + "epoch": 0.651596224373234, + "grad_norm": 1.7610001885426478, + "learning_rate": 1.1436796919701484e-06, + "loss": 0.9864, + "step": 5419 + }, + { + "epoch": 0.651716467263873, + "grad_norm": 1.91526327031633, + "learning_rate": 1.1429757980239115e-06, + "loss": 0.8545, + "step": 5420 + }, + { + "epoch": 0.6518367101545122, + "grad_norm": 3.080156769168779, + "learning_rate": 1.1422720340875636e-06, + "loss": 1.0531, + "step": 5421 + }, + { + "epoch": 0.6519569530451512, + "grad_norm": 2.175298722898534, + "learning_rate": 1.1415684002678671e-06, + "loss": 1.0282, + "step": 5422 + }, + { + "epoch": 0.6520771959357903, + "grad_norm": 2.213329732884309, + "learning_rate": 1.1408648966715617e-06, + "loss": 1.0163, + "step": 5423 + }, + { + "epoch": 0.6521974388264293, + "grad_norm": 1.9057102377230264, + "learning_rate": 1.1401615234053683e-06, + "loss": 0.962, + "step": 5424 + }, + { + "epoch": 0.6523176817170685, + "grad_norm": 1.8181895700202764, + "learning_rate": 1.1394582805759885e-06, + "loss": 0.9967, + "step": 5425 + }, + { + "epoch": 0.6524379246077076, + "grad_norm": 1.8117576262665045, + "learning_rate": 1.1387551682901022e-06, + "loss": 0.995, + "step": 5426 + }, + { + "epoch": 0.6525581674983466, + "grad_norm": 1.845373764119325, + "learning_rate": 1.138052186654373e-06, + "loss": 0.9431, + "step": 5427 + }, + { + "epoch": 0.6526784103889858, + "grad_norm": 2.134032965715616, + "learning_rate": 1.1373493357754417e-06, + "loss": 1.1177, + "step": 5428 + }, + { + "epoch": 0.6527986532796248, + "grad_norm": 1.9638198399159381, + "learning_rate": 1.1366466157599303e-06, + "loss": 1.0075, + "step": 5429 + }, + { + "epoch": 0.6529188961702639, + "grad_norm": 2.29954798859325, + "learning_rate": 1.1359440267144412e-06, + "loss": 0.9974, + "step": 5430 + }, + { + "epoch": 0.653039139060903, + "grad_norm": 3.021677692839979, + "learning_rate": 1.1352415687455556e-06, + "loss": 0.9781, + "step": 5431 + }, + { + "epoch": 0.6531593819515421, + "grad_norm": 2.578567189529008, + "learning_rate": 1.1345392419598362e-06, + "loss": 0.8782, + "step": 5432 + }, + { + "epoch": 0.6532796248421812, + "grad_norm": 2.7686823093931667, + "learning_rate": 1.1338370464638263e-06, + "loss": 0.9554, + "step": 5433 + }, + { + "epoch": 0.6533998677328203, + "grad_norm": 2.580699121099513, + "learning_rate": 1.1331349823640474e-06, + "loss": 0.8758, + "step": 5434 + }, + { + "epoch": 0.6535201106234594, + "grad_norm": 2.6912785072074823, + "learning_rate": 1.132433049767003e-06, + "loss": 1.0154, + "step": 5435 + }, + { + "epoch": 0.6536403535140984, + "grad_norm": 2.0099187705484223, + "learning_rate": 1.1317312487791748e-06, + "loss": 1.0477, + "step": 5436 + }, + { + "epoch": 0.6537605964047376, + "grad_norm": 2.240454559015193, + "learning_rate": 1.1310295795070253e-06, + "loss": 0.9682, + "step": 5437 + }, + { + "epoch": 0.6538808392953767, + "grad_norm": 1.8761139421733781, + "learning_rate": 1.1303280420569982e-06, + "loss": 1.0479, + "step": 5438 + }, + { + "epoch": 0.6540010821860157, + "grad_norm": 1.7291618561582025, + "learning_rate": 1.1296266365355158e-06, + "loss": 1.0112, + "step": 5439 + }, + { + "epoch": 0.6541213250766549, + "grad_norm": 2.003252246165614, + "learning_rate": 1.1289253630489806e-06, + "loss": 0.9741, + "step": 5440 + }, + { + "epoch": 0.6542415679672939, + "grad_norm": 1.9842501623579691, + "learning_rate": 1.1282242217037753e-06, + "loss": 0.9654, + "step": 5441 + }, + { + "epoch": 0.654361810857933, + "grad_norm": 2.666392451333601, + "learning_rate": 1.127523212606262e-06, + "loss": 0.8589, + "step": 5442 + }, + { + "epoch": 0.6544820537485722, + "grad_norm": 1.7300549565665475, + "learning_rate": 1.1268223358627835e-06, + "loss": 0.9689, + "step": 5443 + }, + { + "epoch": 0.6546022966392112, + "grad_norm": 1.8568869862511457, + "learning_rate": 1.126121591579663e-06, + "loss": 0.9575, + "step": 5444 + }, + { + "epoch": 0.6547225395298503, + "grad_norm": 1.8450410493042029, + "learning_rate": 1.1254209798632018e-06, + "loss": 0.9282, + "step": 5445 + }, + { + "epoch": 0.6548427824204894, + "grad_norm": 1.8697188707665877, + "learning_rate": 1.124720500819683e-06, + "loss": 1.0843, + "step": 5446 + }, + { + "epoch": 0.6549630253111285, + "grad_norm": 1.8613458014146236, + "learning_rate": 1.1240201545553682e-06, + "loss": 1.0617, + "step": 5447 + }, + { + "epoch": 0.6550832682017675, + "grad_norm": 1.8279984259191702, + "learning_rate": 1.1233199411764987e-06, + "loss": 0.9712, + "step": 5448 + }, + { + "epoch": 0.6552035110924067, + "grad_norm": 1.8034050418231193, + "learning_rate": 1.1226198607892978e-06, + "loss": 0.926, + "step": 5449 + }, + { + "epoch": 0.6553237539830458, + "grad_norm": 1.8996699849680023, + "learning_rate": 1.1219199134999664e-06, + "loss": 1.0377, + "step": 5450 + }, + { + "epoch": 0.6554439968736848, + "grad_norm": 2.309441786692054, + "learning_rate": 1.1212200994146863e-06, + "loss": 1.0257, + "step": 5451 + }, + { + "epoch": 0.655564239764324, + "grad_norm": 1.894300183234986, + "learning_rate": 1.120520418639618e-06, + "loss": 0.9969, + "step": 5452 + }, + { + "epoch": 0.655684482654963, + "grad_norm": 2.050074282905302, + "learning_rate": 1.119820871280903e-06, + "loss": 1.0717, + "step": 5453 + }, + { + "epoch": 0.6558047255456021, + "grad_norm": 2.2424290756186, + "learning_rate": 1.1191214574446614e-06, + "loss": 0.9688, + "step": 5454 + }, + { + "epoch": 0.6559249684362413, + "grad_norm": 1.5039704544432164, + "learning_rate": 1.118422177236995e-06, + "loss": 1.0357, + "step": 5455 + }, + { + "epoch": 0.6560452113268803, + "grad_norm": 2.5498851002257, + "learning_rate": 1.1177230307639835e-06, + "loss": 1.0937, + "step": 5456 + }, + { + "epoch": 0.6561654542175194, + "grad_norm": 1.8929029880740116, + "learning_rate": 1.1170240181316865e-06, + "loss": 1.0231, + "step": 5457 + }, + { + "epoch": 0.6562856971081584, + "grad_norm": 1.9353713370750911, + "learning_rate": 1.1163251394461442e-06, + "loss": 1.0314, + "step": 5458 + }, + { + "epoch": 0.6564059399987976, + "grad_norm": 1.9343070515914544, + "learning_rate": 1.1156263948133746e-06, + "loss": 1.0618, + "step": 5459 + }, + { + "epoch": 0.6565261828894366, + "grad_norm": 1.7670803992131556, + "learning_rate": 1.1149277843393787e-06, + "loss": 1.014, + "step": 5460 + }, + { + "epoch": 0.6566464257800757, + "grad_norm": 2.463132817629353, + "learning_rate": 1.1142293081301342e-06, + "loss": 0.8705, + "step": 5461 + }, + { + "epoch": 0.6567666686707149, + "grad_norm": 2.0292193312602635, + "learning_rate": 1.1135309662915995e-06, + "loss": 0.915, + "step": 5462 + }, + { + "epoch": 0.6568869115613539, + "grad_norm": 2.1124663130201675, + "learning_rate": 1.112832758929712e-06, + "loss": 0.8391, + "step": 5463 + }, + { + "epoch": 0.657007154451993, + "grad_norm": 2.8376005986736, + "learning_rate": 1.11213468615039e-06, + "loss": 0.9868, + "step": 5464 + }, + { + "epoch": 0.6571273973426321, + "grad_norm": 1.7703407574895813, + "learning_rate": 1.1114367480595292e-06, + "loss": 0.9891, + "step": 5465 + }, + { + "epoch": 0.6572476402332712, + "grad_norm": 6.396673124600653, + "learning_rate": 1.1107389447630086e-06, + "loss": 1.0509, + "step": 5466 + }, + { + "epoch": 0.6573678831239103, + "grad_norm": 2.7193467811588006, + "learning_rate": 1.1100412763666818e-06, + "loss": 1.0212, + "step": 5467 + }, + { + "epoch": 0.6574881260145494, + "grad_norm": 1.5125986934020188, + "learning_rate": 1.1093437429763865e-06, + "loss": 1.039, + "step": 5468 + }, + { + "epoch": 0.6576083689051885, + "grad_norm": 2.0092114276665045, + "learning_rate": 1.1086463446979361e-06, + "loss": 0.9707, + "step": 5469 + }, + { + "epoch": 0.6577286117958275, + "grad_norm": 2.0664660151299397, + "learning_rate": 1.1079490816371277e-06, + "loss": 1.0125, + "step": 5470 + }, + { + "epoch": 0.6578488546864667, + "grad_norm": 2.829022681308841, + "learning_rate": 1.1072519538997352e-06, + "loss": 0.9826, + "step": 5471 + }, + { + "epoch": 0.6579690975771058, + "grad_norm": 1.8832612828451343, + "learning_rate": 1.1065549615915095e-06, + "loss": 1.059, + "step": 5472 + }, + { + "epoch": 0.6580893404677448, + "grad_norm": 2.399927626949093, + "learning_rate": 1.105858104818187e-06, + "loss": 1.0223, + "step": 5473 + }, + { + "epoch": 0.658209583358384, + "grad_norm": 2.4591437690347075, + "learning_rate": 1.105161383685478e-06, + "loss": 0.9899, + "step": 5474 + }, + { + "epoch": 0.658329826249023, + "grad_norm": 0.7346318384241376, + "learning_rate": 1.1044647982990771e-06, + "loss": 0.801, + "step": 5475 + }, + { + "epoch": 0.6584500691396621, + "grad_norm": 2.4848782971243835, + "learning_rate": 1.1037683487646536e-06, + "loss": 0.8821, + "step": 5476 + }, + { + "epoch": 0.6585703120303013, + "grad_norm": 2.1395842535158707, + "learning_rate": 1.1030720351878583e-06, + "loss": 1.0069, + "step": 5477 + }, + { + "epoch": 0.6586905549209403, + "grad_norm": 0.8144689691040742, + "learning_rate": 1.102375857674323e-06, + "loss": 0.8134, + "step": 5478 + }, + { + "epoch": 0.6588107978115794, + "grad_norm": 2.2250000942958854, + "learning_rate": 1.1016798163296561e-06, + "loss": 1.1386, + "step": 5479 + }, + { + "epoch": 0.6589310407022185, + "grad_norm": 2.0409922621270518, + "learning_rate": 1.1009839112594471e-06, + "loss": 0.8975, + "step": 5480 + }, + { + "epoch": 0.6590512835928576, + "grad_norm": 2.183019135947415, + "learning_rate": 1.1002881425692638e-06, + "loss": 0.9567, + "step": 5481 + }, + { + "epoch": 0.6591715264834966, + "grad_norm": 1.7075374036185766, + "learning_rate": 1.0995925103646532e-06, + "loss": 0.9906, + "step": 5482 + }, + { + "epoch": 0.6592917693741358, + "grad_norm": 1.531773691737848, + "learning_rate": 1.0988970147511437e-06, + "loss": 0.9031, + "step": 5483 + }, + { + "epoch": 0.6594120122647749, + "grad_norm": 2.276095275480349, + "learning_rate": 1.0982016558342405e-06, + "loss": 1.0404, + "step": 5484 + }, + { + "epoch": 0.6595322551554139, + "grad_norm": 2.240199999225413, + "learning_rate": 1.0975064337194291e-06, + "loss": 0.947, + "step": 5485 + }, + { + "epoch": 0.6596524980460531, + "grad_norm": 4.920294613895027, + "learning_rate": 1.0968113485121743e-06, + "loss": 0.937, + "step": 5486 + }, + { + "epoch": 0.6597727409366921, + "grad_norm": 1.9899625433458987, + "learning_rate": 1.0961164003179185e-06, + "loss": 1.0374, + "step": 5487 + }, + { + "epoch": 0.6598929838273312, + "grad_norm": 1.9928102006281327, + "learning_rate": 1.0954215892420884e-06, + "loss": 1.0774, + "step": 5488 + }, + { + "epoch": 0.6600132267179702, + "grad_norm": 1.73983605960813, + "learning_rate": 1.094726915390082e-06, + "loss": 0.9426, + "step": 5489 + }, + { + "epoch": 0.6601334696086094, + "grad_norm": 2.1869967290420385, + "learning_rate": 1.0940323788672836e-06, + "loss": 0.9325, + "step": 5490 + }, + { + "epoch": 0.6602537124992485, + "grad_norm": 1.6589915779088542, + "learning_rate": 1.0933379797790522e-06, + "loss": 0.9734, + "step": 5491 + }, + { + "epoch": 0.6603739553898875, + "grad_norm": 2.636071155282359, + "learning_rate": 1.0926437182307293e-06, + "loss": 0.9523, + "step": 5492 + }, + { + "epoch": 0.6604941982805267, + "grad_norm": 1.8016381862387825, + "learning_rate": 1.0919495943276338e-06, + "loss": 1.0187, + "step": 5493 + }, + { + "epoch": 0.6606144411711657, + "grad_norm": 3.466057772225921, + "learning_rate": 1.0912556081750611e-06, + "loss": 1.0009, + "step": 5494 + }, + { + "epoch": 0.6607346840618048, + "grad_norm": 2.203911282527875, + "learning_rate": 1.0905617598782909e-06, + "loss": 1.0, + "step": 5495 + }, + { + "epoch": 0.660854926952444, + "grad_norm": 1.962485929305058, + "learning_rate": 1.0898680495425775e-06, + "loss": 1.0526, + "step": 5496 + }, + { + "epoch": 0.660975169843083, + "grad_norm": 1.7669991066481918, + "learning_rate": 1.0891744772731594e-06, + "loss": 1.0397, + "step": 5497 + }, + { + "epoch": 0.6610954127337221, + "grad_norm": 1.7424416121520494, + "learning_rate": 1.088481043175248e-06, + "loss": 0.8964, + "step": 5498 + }, + { + "epoch": 0.6612156556243612, + "grad_norm": 1.9227885734350316, + "learning_rate": 1.0877877473540368e-06, + "loss": 0.9964, + "step": 5499 + }, + { + "epoch": 0.6613358985150003, + "grad_norm": 2.355150394213502, + "learning_rate": 1.0870945899147002e-06, + "loss": 0.961, + "step": 5500 + }, + { + "epoch": 0.6614561414056394, + "grad_norm": 1.9092728457930306, + "learning_rate": 1.0864015709623879e-06, + "loss": 0.9994, + "step": 5501 + }, + { + "epoch": 0.6615763842962785, + "grad_norm": 2.177346123953115, + "learning_rate": 1.0857086906022313e-06, + "loss": 1.0396, + "step": 5502 + }, + { + "epoch": 0.6616966271869176, + "grad_norm": 1.890706273373382, + "learning_rate": 1.0850159489393388e-06, + "loss": 0.9697, + "step": 5503 + }, + { + "epoch": 0.6618168700775566, + "grad_norm": 2.176535018670731, + "learning_rate": 1.0843233460787992e-06, + "loss": 1.0591, + "step": 5504 + }, + { + "epoch": 0.6619371129681958, + "grad_norm": 1.893599951814426, + "learning_rate": 1.0836308821256805e-06, + "loss": 1.0171, + "step": 5505 + }, + { + "epoch": 0.6620573558588349, + "grad_norm": 2.1241082957186808, + "learning_rate": 1.0829385571850282e-06, + "loss": 1.0153, + "step": 5506 + }, + { + "epoch": 0.6621775987494739, + "grad_norm": 2.5967017234162775, + "learning_rate": 1.0822463713618679e-06, + "loss": 1.0715, + "step": 5507 + }, + { + "epoch": 0.6622978416401131, + "grad_norm": 2.2099411947099243, + "learning_rate": 1.0815543247612034e-06, + "loss": 1.0852, + "step": 5508 + }, + { + "epoch": 0.6624180845307521, + "grad_norm": 1.5950917598434757, + "learning_rate": 1.0808624174880168e-06, + "loss": 1.0675, + "step": 5509 + }, + { + "epoch": 0.6625383274213912, + "grad_norm": 1.8336201428991323, + "learning_rate": 1.080170649647272e-06, + "loss": 1.0373, + "step": 5510 + }, + { + "epoch": 0.6626585703120303, + "grad_norm": 1.6422809781111374, + "learning_rate": 1.0794790213439068e-06, + "loss": 0.9103, + "step": 5511 + }, + { + "epoch": 0.6627788132026694, + "grad_norm": 2.1526908802282483, + "learning_rate": 1.078787532682843e-06, + "loss": 1.0194, + "step": 5512 + }, + { + "epoch": 0.6628990560933085, + "grad_norm": 2.332563023344393, + "learning_rate": 1.0780961837689773e-06, + "loss": 0.9947, + "step": 5513 + }, + { + "epoch": 0.6630192989839476, + "grad_norm": 2.3720118145190114, + "learning_rate": 1.0774049747071883e-06, + "loss": 0.9351, + "step": 5514 + }, + { + "epoch": 0.6631395418745867, + "grad_norm": 2.033824052640643, + "learning_rate": 1.076713905602332e-06, + "loss": 0.9206, + "step": 5515 + }, + { + "epoch": 0.6632597847652257, + "grad_norm": 1.7030757231976834, + "learning_rate": 1.07602297655924e-06, + "loss": 1.0472, + "step": 5516 + }, + { + "epoch": 0.6633800276558649, + "grad_norm": 1.8698448520700823, + "learning_rate": 1.0753321876827292e-06, + "loss": 1.0469, + "step": 5517 + }, + { + "epoch": 0.663500270546504, + "grad_norm": 3.720629049049352, + "learning_rate": 1.0746415390775893e-06, + "loss": 0.9772, + "step": 5518 + }, + { + "epoch": 0.663620513437143, + "grad_norm": 1.948257484471588, + "learning_rate": 1.0739510308485939e-06, + "loss": 1.002, + "step": 5519 + }, + { + "epoch": 0.6637407563277821, + "grad_norm": 0.8165229093441069, + "learning_rate": 1.07326066310049e-06, + "loss": 0.8627, + "step": 5520 + }, + { + "epoch": 0.6638609992184212, + "grad_norm": 2.178740811966012, + "learning_rate": 1.0725704359380059e-06, + "loss": 1.0284, + "step": 5521 + }, + { + "epoch": 0.6639812421090603, + "grad_norm": 1.9864566727624469, + "learning_rate": 1.0718803494658497e-06, + "loss": 0.9549, + "step": 5522 + }, + { + "epoch": 0.6641014849996993, + "grad_norm": 2.3452173344608624, + "learning_rate": 1.071190403788707e-06, + "loss": 1.0804, + "step": 5523 + }, + { + "epoch": 0.6642217278903385, + "grad_norm": 1.8318899136746056, + "learning_rate": 1.0705005990112415e-06, + "loss": 0.9929, + "step": 5524 + }, + { + "epoch": 0.6643419707809776, + "grad_norm": 2.54424154420287, + "learning_rate": 1.0698109352380957e-06, + "loss": 0.98, + "step": 5525 + }, + { + "epoch": 0.6644622136716166, + "grad_norm": 1.8604731802848775, + "learning_rate": 1.0691214125738909e-06, + "loss": 1.0159, + "step": 5526 + }, + { + "epoch": 0.6645824565622558, + "grad_norm": 0.7991635613949629, + "learning_rate": 1.0684320311232287e-06, + "loss": 0.8129, + "step": 5527 + }, + { + "epoch": 0.6647026994528948, + "grad_norm": 2.444759025946379, + "learning_rate": 1.0677427909906865e-06, + "loss": 1.0493, + "step": 5528 + }, + { + "epoch": 0.6648229423435339, + "grad_norm": 2.0062368422222345, + "learning_rate": 1.0670536922808216e-06, + "loss": 0.9578, + "step": 5529 + }, + { + "epoch": 0.6649431852341731, + "grad_norm": 2.246037703105077, + "learning_rate": 1.06636473509817e-06, + "loss": 0.9576, + "step": 5530 + }, + { + "epoch": 0.6650634281248121, + "grad_norm": 2.0919860697295243, + "learning_rate": 1.0656759195472447e-06, + "loss": 1.0403, + "step": 5531 + }, + { + "epoch": 0.6651836710154512, + "grad_norm": 0.7749117278003788, + "learning_rate": 1.0649872457325414e-06, + "loss": 0.8353, + "step": 5532 + }, + { + "epoch": 0.6653039139060903, + "grad_norm": 0.8610637371472171, + "learning_rate": 1.0642987137585278e-06, + "loss": 0.7882, + "step": 5533 + }, + { + "epoch": 0.6654241567967294, + "grad_norm": 1.8168844926531666, + "learning_rate": 1.0636103237296561e-06, + "loss": 1.0639, + "step": 5534 + }, + { + "epoch": 0.6655443996873684, + "grad_norm": 2.171865120186049, + "learning_rate": 1.062922075750353e-06, + "loss": 1.0797, + "step": 5535 + }, + { + "epoch": 0.6656646425780076, + "grad_norm": 2.0553063450634568, + "learning_rate": 1.0622339699250267e-06, + "loss": 0.9571, + "step": 5536 + }, + { + "epoch": 0.6657848854686467, + "grad_norm": 1.720656707172226, + "learning_rate": 1.0615460063580624e-06, + "loss": 1.0298, + "step": 5537 + }, + { + "epoch": 0.6659051283592857, + "grad_norm": 1.8665677243077707, + "learning_rate": 1.060858185153821e-06, + "loss": 0.9674, + "step": 5538 + }, + { + "epoch": 0.6660253712499249, + "grad_norm": 2.3143236341346913, + "learning_rate": 1.0601705064166474e-06, + "loss": 0.9974, + "step": 5539 + }, + { + "epoch": 0.666145614140564, + "grad_norm": 3.0588963384507175, + "learning_rate": 1.0594829702508596e-06, + "loss": 0.9747, + "step": 5540 + }, + { + "epoch": 0.666265857031203, + "grad_norm": 2.008787994310302, + "learning_rate": 1.0587955767607592e-06, + "loss": 0.7845, + "step": 5541 + }, + { + "epoch": 0.6663860999218422, + "grad_norm": 3.4298808884027405, + "learning_rate": 1.0581083260506206e-06, + "loss": 1.007, + "step": 5542 + }, + { + "epoch": 0.6665063428124812, + "grad_norm": 2.0177377672728554, + "learning_rate": 1.0574212182246993e-06, + "loss": 1.0046, + "step": 5543 + }, + { + "epoch": 0.6666265857031203, + "grad_norm": 2.5554543373776735, + "learning_rate": 1.0567342533872303e-06, + "loss": 0.9934, + "step": 5544 + }, + { + "epoch": 0.6667468285937594, + "grad_norm": 1.765276308990329, + "learning_rate": 1.0560474316424255e-06, + "loss": 1.0466, + "step": 5545 + }, + { + "epoch": 0.6668670714843985, + "grad_norm": 2.371519449597008, + "learning_rate": 1.0553607530944746e-06, + "loss": 0.9741, + "step": 5546 + }, + { + "epoch": 0.6669873143750376, + "grad_norm": 2.8425558748012656, + "learning_rate": 1.0546742178475463e-06, + "loss": 1.1305, + "step": 5547 + }, + { + "epoch": 0.6671075572656767, + "grad_norm": 1.808805449188811, + "learning_rate": 1.0539878260057868e-06, + "loss": 1.1039, + "step": 5548 + }, + { + "epoch": 0.6672278001563158, + "grad_norm": 2.7588875812055083, + "learning_rate": 1.0533015776733226e-06, + "loss": 0.921, + "step": 5549 + }, + { + "epoch": 0.6673480430469548, + "grad_norm": 2.4585049671668315, + "learning_rate": 1.0526154729542566e-06, + "loss": 1.022, + "step": 5550 + }, + { + "epoch": 0.6674682859375939, + "grad_norm": 2.5311272673816427, + "learning_rate": 1.0519295119526699e-06, + "loss": 1.0377, + "step": 5551 + }, + { + "epoch": 0.667588528828233, + "grad_norm": 1.7020066815613826, + "learning_rate": 1.0512436947726227e-06, + "loss": 1.0693, + "step": 5552 + }, + { + "epoch": 0.6677087717188721, + "grad_norm": 2.372407200629383, + "learning_rate": 1.0505580215181517e-06, + "loss": 0.8899, + "step": 5553 + }, + { + "epoch": 0.6678290146095112, + "grad_norm": 0.7911597226606056, + "learning_rate": 1.0498724922932753e-06, + "loss": 0.8031, + "step": 5554 + }, + { + "epoch": 0.6679492575001503, + "grad_norm": 2.5969505326616824, + "learning_rate": 1.0491871072019851e-06, + "loss": 1.1002, + "step": 5555 + }, + { + "epoch": 0.6680695003907894, + "grad_norm": 2.037142847504842, + "learning_rate": 1.0485018663482555e-06, + "loss": 0.8766, + "step": 5556 + }, + { + "epoch": 0.6681897432814284, + "grad_norm": 2.7407305289365596, + "learning_rate": 1.0478167698360354e-06, + "loss": 0.9448, + "step": 5557 + }, + { + "epoch": 0.6683099861720676, + "grad_norm": 2.725499686527182, + "learning_rate": 1.0471318177692556e-06, + "loss": 0.9364, + "step": 5558 + }, + { + "epoch": 0.6684302290627067, + "grad_norm": 2.1330391899618455, + "learning_rate": 1.046447010251821e-06, + "loss": 0.9987, + "step": 5559 + }, + { + "epoch": 0.6685504719533457, + "grad_norm": 1.6941738259983106, + "learning_rate": 1.0457623473876157e-06, + "loss": 0.9974, + "step": 5560 + }, + { + "epoch": 0.6686707148439849, + "grad_norm": 1.8715969993735202, + "learning_rate": 1.0450778292805046e-06, + "loss": 0.9474, + "step": 5561 + }, + { + "epoch": 0.6687909577346239, + "grad_norm": 1.711944283862236, + "learning_rate": 1.0443934560343267e-06, + "loss": 1.0234, + "step": 5562 + }, + { + "epoch": 0.668911200625263, + "grad_norm": 2.2997734829026864, + "learning_rate": 1.0437092277529034e-06, + "loss": 1.0171, + "step": 5563 + }, + { + "epoch": 0.6690314435159022, + "grad_norm": 2.0966891256119085, + "learning_rate": 1.0430251445400292e-06, + "loss": 0.9695, + "step": 5564 + }, + { + "epoch": 0.6691516864065412, + "grad_norm": 2.370304787107126, + "learning_rate": 1.0423412064994787e-06, + "loss": 0.8608, + "step": 5565 + }, + { + "epoch": 0.6692719292971803, + "grad_norm": 1.9742887785688448, + "learning_rate": 1.0416574137350064e-06, + "loss": 0.9793, + "step": 5566 + }, + { + "epoch": 0.6693921721878194, + "grad_norm": 2.812771508569035, + "learning_rate": 1.0409737663503428e-06, + "loss": 1.0445, + "step": 5567 + }, + { + "epoch": 0.6695124150784585, + "grad_norm": 1.7814880345895856, + "learning_rate": 1.040290264449196e-06, + "loss": 1.064, + "step": 5568 + }, + { + "epoch": 0.6696326579690975, + "grad_norm": 1.9743887667705804, + "learning_rate": 1.0396069081352532e-06, + "loss": 0.878, + "step": 5569 + }, + { + "epoch": 0.6697529008597367, + "grad_norm": 0.7846018793632487, + "learning_rate": 1.0389236975121782e-06, + "loss": 0.7972, + "step": 5570 + }, + { + "epoch": 0.6698731437503758, + "grad_norm": 2.200007403968143, + "learning_rate": 1.0382406326836147e-06, + "loss": 0.9505, + "step": 5571 + }, + { + "epoch": 0.6699933866410148, + "grad_norm": 2.049813410943503, + "learning_rate": 1.0375577137531828e-06, + "loss": 0.9912, + "step": 5572 + }, + { + "epoch": 0.670113629531654, + "grad_norm": 2.1268707343503475, + "learning_rate": 1.0368749408244802e-06, + "loss": 0.9596, + "step": 5573 + }, + { + "epoch": 0.670233872422293, + "grad_norm": 2.176632069326869, + "learning_rate": 1.0361923140010836e-06, + "loss": 1.0239, + "step": 5574 + }, + { + "epoch": 0.6703541153129321, + "grad_norm": 3.839536885508228, + "learning_rate": 1.0355098333865455e-06, + "loss": 0.8691, + "step": 5575 + }, + { + "epoch": 0.6704743582035713, + "grad_norm": 1.7030079652663819, + "learning_rate": 1.0348274990844006e-06, + "loss": 0.9307, + "step": 5576 + }, + { + "epoch": 0.6705946010942103, + "grad_norm": 1.7326039387404204, + "learning_rate": 1.034145311198155e-06, + "loss": 0.9635, + "step": 5577 + }, + { + "epoch": 0.6707148439848494, + "grad_norm": 2.010338763636636, + "learning_rate": 1.0334632698312989e-06, + "loss": 0.8765, + "step": 5578 + }, + { + "epoch": 0.6708350868754885, + "grad_norm": 2.3492629254789223, + "learning_rate": 1.032781375087295e-06, + "loss": 0.9922, + "step": 5579 + }, + { + "epoch": 0.6709553297661276, + "grad_norm": 1.5072486890646997, + "learning_rate": 1.0320996270695891e-06, + "loss": 0.9095, + "step": 5580 + }, + { + "epoch": 0.6710755726567667, + "grad_norm": 2.073502756455914, + "learning_rate": 1.0314180258815998e-06, + "loss": 0.9677, + "step": 5581 + }, + { + "epoch": 0.6711958155474057, + "grad_norm": 2.0928962447197685, + "learning_rate": 1.0307365716267247e-06, + "loss": 0.9791, + "step": 5582 + }, + { + "epoch": 0.6713160584380449, + "grad_norm": 1.9710112407408307, + "learning_rate": 1.0300552644083423e-06, + "loss": 1.0182, + "step": 5583 + }, + { + "epoch": 0.6714363013286839, + "grad_norm": 2.301214258350511, + "learning_rate": 1.0293741043298036e-06, + "loss": 0.9605, + "step": 5584 + }, + { + "epoch": 0.671556544219323, + "grad_norm": 5.007689285547858, + "learning_rate": 1.0286930914944436e-06, + "loss": 0.953, + "step": 5585 + }, + { + "epoch": 0.6716767871099621, + "grad_norm": 2.997005716760712, + "learning_rate": 1.0280122260055684e-06, + "loss": 1.0146, + "step": 5586 + }, + { + "epoch": 0.6717970300006012, + "grad_norm": 2.032959556098492, + "learning_rate": 1.0273315079664652e-06, + "loss": 1.06, + "step": 5587 + }, + { + "epoch": 0.6719172728912403, + "grad_norm": 2.3091037658068614, + "learning_rate": 1.0266509374803992e-06, + "loss": 0.9802, + "step": 5588 + }, + { + "epoch": 0.6720375157818794, + "grad_norm": 2.542345008589549, + "learning_rate": 1.0259705146506123e-06, + "loss": 1.0801, + "step": 5589 + }, + { + "epoch": 0.6721577586725185, + "grad_norm": 2.1588941201507508, + "learning_rate": 1.025290239580324e-06, + "loss": 1.012, + "step": 5590 + }, + { + "epoch": 0.6722780015631575, + "grad_norm": 1.815419575138256, + "learning_rate": 1.0246101123727313e-06, + "loss": 0.9913, + "step": 5591 + }, + { + "epoch": 0.6723982444537967, + "grad_norm": 1.8755322654550142, + "learning_rate": 1.0239301331310085e-06, + "loss": 1.0264, + "step": 5592 + }, + { + "epoch": 0.6725184873444358, + "grad_norm": 1.748794890039592, + "learning_rate": 1.0232503019583088e-06, + "loss": 1.1225, + "step": 5593 + }, + { + "epoch": 0.6726387302350748, + "grad_norm": 2.5176670481956243, + "learning_rate": 1.0225706189577619e-06, + "loss": 0.9351, + "step": 5594 + }, + { + "epoch": 0.672758973125714, + "grad_norm": 2.6546301895610354, + "learning_rate": 1.021891084232475e-06, + "loss": 0.9812, + "step": 5595 + }, + { + "epoch": 0.672879216016353, + "grad_norm": 2.2925781749180008, + "learning_rate": 1.0212116978855325e-06, + "loss": 1.0352, + "step": 5596 + }, + { + "epoch": 0.6729994589069921, + "grad_norm": 1.7670482199479147, + "learning_rate": 1.020532460019997e-06, + "loss": 1.0247, + "step": 5597 + }, + { + "epoch": 0.6731197017976313, + "grad_norm": 1.8165960397319554, + "learning_rate": 1.0198533707389096e-06, + "loss": 0.9475, + "step": 5598 + }, + { + "epoch": 0.6732399446882703, + "grad_norm": 1.8776088367109558, + "learning_rate": 1.0191744301452853e-06, + "loss": 0.9687, + "step": 5599 + }, + { + "epoch": 0.6733601875789094, + "grad_norm": 1.754891371666314, + "learning_rate": 1.0184956383421208e-06, + "loss": 0.9392, + "step": 5600 + }, + { + "epoch": 0.6734804304695485, + "grad_norm": 3.3483299191807046, + "learning_rate": 1.017816995432387e-06, + "loss": 0.8893, + "step": 5601 + }, + { + "epoch": 0.6736006733601876, + "grad_norm": 2.377196450736259, + "learning_rate": 1.0171385015190353e-06, + "loss": 0.9826, + "step": 5602 + }, + { + "epoch": 0.6737209162508266, + "grad_norm": 1.9459612274071079, + "learning_rate": 1.0164601567049908e-06, + "loss": 0.9676, + "step": 5603 + }, + { + "epoch": 0.6738411591414658, + "grad_norm": 1.7488449917257471, + "learning_rate": 1.015781961093158e-06, + "loss": 1.0361, + "step": 5604 + }, + { + "epoch": 0.6739614020321049, + "grad_norm": 1.8010055170647197, + "learning_rate": 1.0151039147864197e-06, + "loss": 1.0073, + "step": 5605 + }, + { + "epoch": 0.6740816449227439, + "grad_norm": 19.665085982254265, + "learning_rate": 1.0144260178876336e-06, + "loss": 0.902, + "step": 5606 + }, + { + "epoch": 0.6742018878133831, + "grad_norm": 2.2998989207406004, + "learning_rate": 1.0137482704996388e-06, + "loss": 0.9074, + "step": 5607 + }, + { + "epoch": 0.6743221307040221, + "grad_norm": 1.912710218158216, + "learning_rate": 1.0130706727252461e-06, + "loss": 1.0267, + "step": 5608 + }, + { + "epoch": 0.6744423735946612, + "grad_norm": 2.8840736173844737, + "learning_rate": 1.0123932246672468e-06, + "loss": 0.9197, + "step": 5609 + }, + { + "epoch": 0.6745626164853004, + "grad_norm": 0.7548176291922347, + "learning_rate": 1.0117159264284114e-06, + "loss": 0.7954, + "step": 5610 + }, + { + "epoch": 0.6746828593759394, + "grad_norm": 2.5568090373174344, + "learning_rate": 1.0110387781114837e-06, + "loss": 1.0064, + "step": 5611 + }, + { + "epoch": 0.6748031022665785, + "grad_norm": 2.1065546466322793, + "learning_rate": 1.0103617798191872e-06, + "loss": 1.0107, + "step": 5612 + }, + { + "epoch": 0.6749233451572175, + "grad_norm": 3.2210508844714023, + "learning_rate": 1.0096849316542217e-06, + "loss": 1.0638, + "step": 5613 + }, + { + "epoch": 0.6750435880478567, + "grad_norm": 4.079589580761885, + "learning_rate": 1.0090082337192643e-06, + "loss": 0.9817, + "step": 5614 + }, + { + "epoch": 0.6751638309384957, + "grad_norm": 2.1576457274762766, + "learning_rate": 1.0083316861169705e-06, + "loss": 1.0217, + "step": 5615 + }, + { + "epoch": 0.6752840738291348, + "grad_norm": 2.147768006482071, + "learning_rate": 1.0076552889499713e-06, + "loss": 0.9475, + "step": 5616 + }, + { + "epoch": 0.675404316719774, + "grad_norm": 1.901407814473507, + "learning_rate": 1.006979042320876e-06, + "loss": 0.9724, + "step": 5617 + }, + { + "epoch": 0.675524559610413, + "grad_norm": 2.209584499500055, + "learning_rate": 1.0063029463322702e-06, + "loss": 0.8693, + "step": 5618 + }, + { + "epoch": 0.6756448025010521, + "grad_norm": 2.8683554118265553, + "learning_rate": 1.0056270010867164e-06, + "loss": 0.9919, + "step": 5619 + }, + { + "epoch": 0.6757650453916912, + "grad_norm": 2.4329817539695866, + "learning_rate": 1.004951206686758e-06, + "loss": 1.018, + "step": 5620 + }, + { + "epoch": 0.6758852882823303, + "grad_norm": 1.878037408446004, + "learning_rate": 1.0042755632349087e-06, + "loss": 0.954, + "step": 5621 + }, + { + "epoch": 0.6760055311729694, + "grad_norm": 3.1108768995620055, + "learning_rate": 1.0036000708336653e-06, + "loss": 0.8692, + "step": 5622 + }, + { + "epoch": 0.6761257740636085, + "grad_norm": 2.246267083155609, + "learning_rate": 1.0029247295854984e-06, + "loss": 1.0352, + "step": 5623 + }, + { + "epoch": 0.6762460169542476, + "grad_norm": 3.0070566788289743, + "learning_rate": 1.0022495395928588e-06, + "loss": 0.9543, + "step": 5624 + }, + { + "epoch": 0.6763662598448866, + "grad_norm": 0.7928130650020235, + "learning_rate": 1.0015745009581697e-06, + "loss": 0.8603, + "step": 5625 + }, + { + "epoch": 0.6764865027355258, + "grad_norm": 1.7754953687129935, + "learning_rate": 1.0008996137838343e-06, + "loss": 0.9075, + "step": 5626 + }, + { + "epoch": 0.6766067456261649, + "grad_norm": 2.079685285124255, + "learning_rate": 1.000224878172234e-06, + "loss": 1.0358, + "step": 5627 + }, + { + "epoch": 0.6767269885168039, + "grad_norm": 2.0869375967466883, + "learning_rate": 9.99550294225724e-07, + "loss": 0.9636, + "step": 5628 + }, + { + "epoch": 0.6768472314074431, + "grad_norm": 1.9636959416690545, + "learning_rate": 9.988758620466402e-07, + "loss": 0.9666, + "step": 5629 + }, + { + "epoch": 0.6769674742980821, + "grad_norm": 1.5835384771395808, + "learning_rate": 9.982015817372917e-07, + "loss": 0.9989, + "step": 5630 + }, + { + "epoch": 0.6770877171887212, + "grad_norm": 1.916650260633936, + "learning_rate": 9.975274533999657e-07, + "loss": 1.06, + "step": 5631 + }, + { + "epoch": 0.6772079600793603, + "grad_norm": 2.75548699645752, + "learning_rate": 9.96853477136929e-07, + "loss": 1.0797, + "step": 5632 + }, + { + "epoch": 0.6773282029699994, + "grad_norm": 2.095331761714691, + "learning_rate": 9.96179653050422e-07, + "loss": 0.9879, + "step": 5633 + }, + { + "epoch": 0.6774484458606385, + "grad_norm": 2.3567320312902718, + "learning_rate": 9.955059812426635e-07, + "loss": 0.9774, + "step": 5634 + }, + { + "epoch": 0.6775686887512776, + "grad_norm": 2.0311026886562025, + "learning_rate": 9.948324618158493e-07, + "loss": 1.0644, + "step": 5635 + }, + { + "epoch": 0.6776889316419167, + "grad_norm": 3.0950670617746434, + "learning_rate": 9.941590948721502e-07, + "loss": 1.0142, + "step": 5636 + }, + { + "epoch": 0.6778091745325557, + "grad_norm": 1.8121202005176884, + "learning_rate": 9.934858805137188e-07, + "loss": 1.0017, + "step": 5637 + }, + { + "epoch": 0.6779294174231949, + "grad_norm": 1.6239827713545776, + "learning_rate": 9.92812818842677e-07, + "loss": 1.0412, + "step": 5638 + }, + { + "epoch": 0.678049660313834, + "grad_norm": 2.1261242529157767, + "learning_rate": 9.921399099611306e-07, + "loss": 0.8776, + "step": 5639 + }, + { + "epoch": 0.678169903204473, + "grad_norm": 1.6966533368579426, + "learning_rate": 9.914671539711588e-07, + "loss": 0.9259, + "step": 5640 + }, + { + "epoch": 0.6782901460951122, + "grad_norm": 3.0904415434878914, + "learning_rate": 9.90794550974817e-07, + "loss": 1.0187, + "step": 5641 + }, + { + "epoch": 0.6784103889857512, + "grad_norm": 2.9811275049151327, + "learning_rate": 9.901221010741407e-07, + "loss": 1.0543, + "step": 5642 + }, + { + "epoch": 0.6785306318763903, + "grad_norm": 1.8732285714750885, + "learning_rate": 9.894498043711375e-07, + "loss": 0.9852, + "step": 5643 + }, + { + "epoch": 0.6786508747670293, + "grad_norm": 2.3262775438780134, + "learning_rate": 9.887776609677962e-07, + "loss": 0.9289, + "step": 5644 + }, + { + "epoch": 0.6787711176576685, + "grad_norm": 1.7279630478201662, + "learning_rate": 9.88105670966079e-07, + "loss": 0.9591, + "step": 5645 + }, + { + "epoch": 0.6788913605483076, + "grad_norm": 1.8671558968012272, + "learning_rate": 9.874338344679283e-07, + "loss": 1.0207, + "step": 5646 + }, + { + "epoch": 0.6790116034389466, + "grad_norm": 1.8390791614331945, + "learning_rate": 9.86762151575259e-07, + "loss": 0.9777, + "step": 5647 + }, + { + "epoch": 0.6791318463295858, + "grad_norm": 1.5001822996624308, + "learning_rate": 9.860906223899651e-07, + "loss": 1.0402, + "step": 5648 + }, + { + "epoch": 0.6792520892202248, + "grad_norm": 1.6789897800291282, + "learning_rate": 9.854192470139184e-07, + "loss": 0.993, + "step": 5649 + }, + { + "epoch": 0.6793723321108639, + "grad_norm": 2.150019614551584, + "learning_rate": 9.847480255489645e-07, + "loss": 0.9574, + "step": 5650 + }, + { + "epoch": 0.6794925750015031, + "grad_norm": 1.9017400704118732, + "learning_rate": 9.840769580969295e-07, + "loss": 0.9264, + "step": 5651 + }, + { + "epoch": 0.6796128178921421, + "grad_norm": 1.9294438382031942, + "learning_rate": 9.834060447596114e-07, + "loss": 1.039, + "step": 5652 + }, + { + "epoch": 0.6797330607827812, + "grad_norm": 2.393295193087577, + "learning_rate": 9.827352856387868e-07, + "loss": 1.0203, + "step": 5653 + }, + { + "epoch": 0.6798533036734203, + "grad_norm": 0.7864392554051262, + "learning_rate": 9.820646808362118e-07, + "loss": 0.8801, + "step": 5654 + }, + { + "epoch": 0.6799735465640594, + "grad_norm": 2.126288416098879, + "learning_rate": 9.813942304536154e-07, + "loss": 0.9668, + "step": 5655 + }, + { + "epoch": 0.6800937894546984, + "grad_norm": 1.7872490499976574, + "learning_rate": 9.807239345927043e-07, + "loss": 0.8751, + "step": 5656 + }, + { + "epoch": 0.6802140323453376, + "grad_norm": 3.1212045127111514, + "learning_rate": 9.80053793355162e-07, + "loss": 0.9553, + "step": 5657 + }, + { + "epoch": 0.6803342752359767, + "grad_norm": 2.305711835392943, + "learning_rate": 9.793838068426472e-07, + "loss": 0.9877, + "step": 5658 + }, + { + "epoch": 0.6804545181266157, + "grad_norm": 2.111560282969274, + "learning_rate": 9.78713975156799e-07, + "loss": 0.8483, + "step": 5659 + }, + { + "epoch": 0.6805747610172549, + "grad_norm": 1.7943031813792685, + "learning_rate": 9.780442983992273e-07, + "loss": 0.9537, + "step": 5660 + }, + { + "epoch": 0.680695003907894, + "grad_norm": 2.169665633075886, + "learning_rate": 9.773747766715238e-07, + "loss": 0.9557, + "step": 5661 + }, + { + "epoch": 0.680815246798533, + "grad_norm": 1.8936143681769557, + "learning_rate": 9.767054100752536e-07, + "loss": 1.041, + "step": 5662 + }, + { + "epoch": 0.6809354896891722, + "grad_norm": 1.955074222648453, + "learning_rate": 9.760361987119584e-07, + "loss": 1.0541, + "step": 5663 + }, + { + "epoch": 0.6810557325798112, + "grad_norm": 2.0317693193187676, + "learning_rate": 9.753671426831592e-07, + "loss": 0.9121, + "step": 5664 + }, + { + "epoch": 0.6811759754704503, + "grad_norm": 1.8416815324186229, + "learning_rate": 9.746982420903483e-07, + "loss": 1.0306, + "step": 5665 + }, + { + "epoch": 0.6812962183610894, + "grad_norm": 1.758750427940116, + "learning_rate": 9.740294970349993e-07, + "loss": 0.9857, + "step": 5666 + }, + { + "epoch": 0.6814164612517285, + "grad_norm": 0.8909712335949913, + "learning_rate": 9.733609076185594e-07, + "loss": 0.8463, + "step": 5667 + }, + { + "epoch": 0.6815367041423676, + "grad_norm": 1.92358551819352, + "learning_rate": 9.72692473942455e-07, + "loss": 1.0756, + "step": 5668 + }, + { + "epoch": 0.6816569470330067, + "grad_norm": 1.7304411944053233, + "learning_rate": 9.720241961080849e-07, + "loss": 1.0159, + "step": 5669 + }, + { + "epoch": 0.6817771899236458, + "grad_norm": 2.079004546160037, + "learning_rate": 9.713560742168259e-07, + "loss": 0.9692, + "step": 5670 + }, + { + "epoch": 0.6818974328142848, + "grad_norm": 2.029189252890881, + "learning_rate": 9.706881083700333e-07, + "loss": 0.9524, + "step": 5671 + }, + { + "epoch": 0.682017675704924, + "grad_norm": 2.0527924475907793, + "learning_rate": 9.700202986690357e-07, + "loss": 1.0634, + "step": 5672 + }, + { + "epoch": 0.682137918595563, + "grad_norm": 1.9451197026934863, + "learning_rate": 9.693526452151413e-07, + "loss": 0.9019, + "step": 5673 + }, + { + "epoch": 0.6822581614862021, + "grad_norm": 2.1050216398236805, + "learning_rate": 9.686851481096305e-07, + "loss": 0.9913, + "step": 5674 + }, + { + "epoch": 0.6823784043768413, + "grad_norm": 2.171214819816944, + "learning_rate": 9.68017807453762e-07, + "loss": 0.9573, + "step": 5675 + }, + { + "epoch": 0.6824986472674803, + "grad_norm": 1.718807566285608, + "learning_rate": 9.673506233487721e-07, + "loss": 0.9705, + "step": 5676 + }, + { + "epoch": 0.6826188901581194, + "grad_norm": 1.9059322279922717, + "learning_rate": 9.666835958958717e-07, + "loss": 1.0994, + "step": 5677 + }, + { + "epoch": 0.6827391330487584, + "grad_norm": 2.1067964971147846, + "learning_rate": 9.660167251962484e-07, + "loss": 1.0414, + "step": 5678 + }, + { + "epoch": 0.6828593759393976, + "grad_norm": 1.6291268105603305, + "learning_rate": 9.653500113510654e-07, + "loss": 1.0159, + "step": 5679 + }, + { + "epoch": 0.6829796188300367, + "grad_norm": 3.5801993275139647, + "learning_rate": 9.646834544614627e-07, + "loss": 0.9057, + "step": 5680 + }, + { + "epoch": 0.6830998617206757, + "grad_norm": 1.8798836368707608, + "learning_rate": 9.64017054628558e-07, + "loss": 1.0012, + "step": 5681 + }, + { + "epoch": 0.6832201046113149, + "grad_norm": 1.6752363265981978, + "learning_rate": 9.63350811953441e-07, + "loss": 1.0266, + "step": 5682 + }, + { + "epoch": 0.6833403475019539, + "grad_norm": 2.366780966922318, + "learning_rate": 9.626847265371826e-07, + "loss": 0.9409, + "step": 5683 + }, + { + "epoch": 0.683460590392593, + "grad_norm": 2.3243667683631277, + "learning_rate": 9.620187984808262e-07, + "loss": 1.0229, + "step": 5684 + }, + { + "epoch": 0.6835808332832322, + "grad_norm": 1.6724709135072304, + "learning_rate": 9.613530278853919e-07, + "loss": 1.0945, + "step": 5685 + }, + { + "epoch": 0.6837010761738712, + "grad_norm": 1.9000089444401667, + "learning_rate": 9.60687414851879e-07, + "loss": 0.981, + "step": 5686 + }, + { + "epoch": 0.6838213190645103, + "grad_norm": 3.9056645069025726, + "learning_rate": 9.600219594812575e-07, + "loss": 1.0106, + "step": 5687 + }, + { + "epoch": 0.6839415619551494, + "grad_norm": 1.7317559339604207, + "learning_rate": 9.593566618744786e-07, + "loss": 0.9666, + "step": 5688 + }, + { + "epoch": 0.6840618048457885, + "grad_norm": 1.8676393991554445, + "learning_rate": 9.58691522132466e-07, + "loss": 0.9752, + "step": 5689 + }, + { + "epoch": 0.6841820477364275, + "grad_norm": 2.0156057194223806, + "learning_rate": 9.58026540356123e-07, + "loss": 1.0844, + "step": 5690 + }, + { + "epoch": 0.6843022906270667, + "grad_norm": 1.7078290598241839, + "learning_rate": 9.573617166463246e-07, + "loss": 1.1091, + "step": 5691 + }, + { + "epoch": 0.6844225335177058, + "grad_norm": 2.041966734134165, + "learning_rate": 9.56697051103924e-07, + "loss": 0.8386, + "step": 5692 + }, + { + "epoch": 0.6845427764083448, + "grad_norm": 2.1229351333648547, + "learning_rate": 9.560325438297522e-07, + "loss": 1.0529, + "step": 5693 + }, + { + "epoch": 0.684663019298984, + "grad_norm": 1.9486755959802906, + "learning_rate": 9.553681949246127e-07, + "loss": 1.1071, + "step": 5694 + }, + { + "epoch": 0.684783262189623, + "grad_norm": 2.0309266053033896, + "learning_rate": 9.547040044892886e-07, + "loss": 0.988, + "step": 5695 + }, + { + "epoch": 0.6849035050802621, + "grad_norm": 0.8842903528041223, + "learning_rate": 9.540399726245354e-07, + "loss": 0.8374, + "step": 5696 + }, + { + "epoch": 0.6850237479709013, + "grad_norm": 2.034937167685583, + "learning_rate": 9.533760994310859e-07, + "loss": 0.9231, + "step": 5697 + }, + { + "epoch": 0.6851439908615403, + "grad_norm": 2.099233932638969, + "learning_rate": 9.527123850096508e-07, + "loss": 0.9893, + "step": 5698 + }, + { + "epoch": 0.6852642337521794, + "grad_norm": 1.841912404519678, + "learning_rate": 9.520488294609142e-07, + "loss": 0.9524, + "step": 5699 + }, + { + "epoch": 0.6853844766428185, + "grad_norm": 0.7569456353212218, + "learning_rate": 9.513854328855368e-07, + "loss": 0.7755, + "step": 5700 + }, + { + "epoch": 0.6855047195334576, + "grad_norm": 2.0305240214076936, + "learning_rate": 9.507221953841558e-07, + "loss": 1.0489, + "step": 5701 + }, + { + "epoch": 0.6856249624240967, + "grad_norm": 1.6959854407675232, + "learning_rate": 9.500591170573824e-07, + "loss": 1.0154, + "step": 5702 + }, + { + "epoch": 0.6857452053147358, + "grad_norm": 2.2808801011445383, + "learning_rate": 9.493961980058078e-07, + "loss": 0.9809, + "step": 5703 + }, + { + "epoch": 0.6858654482053749, + "grad_norm": 8.75992217843661, + "learning_rate": 9.48733438329993e-07, + "loss": 0.9099, + "step": 5704 + }, + { + "epoch": 0.6859856910960139, + "grad_norm": 1.9165797282971124, + "learning_rate": 9.480708381304807e-07, + "loss": 0.9783, + "step": 5705 + }, + { + "epoch": 0.6861059339866531, + "grad_norm": 2.1207614589221047, + "learning_rate": 9.474083975077858e-07, + "loss": 1.0742, + "step": 5706 + }, + { + "epoch": 0.6862261768772921, + "grad_norm": 2.2098590930189497, + "learning_rate": 9.467461165623994e-07, + "loss": 1.0361, + "step": 5707 + }, + { + "epoch": 0.6863464197679312, + "grad_norm": 8.59327235888543, + "learning_rate": 9.46083995394791e-07, + "loss": 1.0341, + "step": 5708 + }, + { + "epoch": 0.6864666626585703, + "grad_norm": 1.9199182737361316, + "learning_rate": 9.454220341054012e-07, + "loss": 0.8744, + "step": 5709 + }, + { + "epoch": 0.6865869055492094, + "grad_norm": 1.9324876492242575, + "learning_rate": 9.447602327946512e-07, + "loss": 1.0414, + "step": 5710 + }, + { + "epoch": 0.6867071484398485, + "grad_norm": 1.981556971502721, + "learning_rate": 9.440985915629338e-07, + "loss": 1.0043, + "step": 5711 + }, + { + "epoch": 0.6868273913304875, + "grad_norm": 1.9937778482913793, + "learning_rate": 9.434371105106223e-07, + "loss": 0.9686, + "step": 5712 + }, + { + "epoch": 0.6869476342211267, + "grad_norm": 1.9574054657189865, + "learning_rate": 9.427757897380602e-07, + "loss": 0.9478, + "step": 5713 + }, + { + "epoch": 0.6870678771117658, + "grad_norm": 2.3277172781285516, + "learning_rate": 9.421146293455695e-07, + "loss": 1.0833, + "step": 5714 + }, + { + "epoch": 0.6871881200024048, + "grad_norm": 2.159143248293265, + "learning_rate": 9.414536294334489e-07, + "loss": 0.925, + "step": 5715 + }, + { + "epoch": 0.687308362893044, + "grad_norm": 3.0103622132406636, + "learning_rate": 9.407927901019708e-07, + "loss": 0.9323, + "step": 5716 + }, + { + "epoch": 0.687428605783683, + "grad_norm": 1.9897962992132312, + "learning_rate": 9.401321114513854e-07, + "loss": 1.0055, + "step": 5717 + }, + { + "epoch": 0.6875488486743221, + "grad_norm": 1.7829093565779481, + "learning_rate": 9.394715935819155e-07, + "loss": 0.9919, + "step": 5718 + }, + { + "epoch": 0.6876690915649613, + "grad_norm": 2.12376098414427, + "learning_rate": 9.388112365937608e-07, + "loss": 0.86, + "step": 5719 + }, + { + "epoch": 0.6877893344556003, + "grad_norm": 2.1591383896859018, + "learning_rate": 9.381510405870985e-07, + "loss": 1.0615, + "step": 5720 + }, + { + "epoch": 0.6879095773462394, + "grad_norm": 2.914336408980939, + "learning_rate": 9.374910056620791e-07, + "loss": 1.0104, + "step": 5721 + }, + { + "epoch": 0.6880298202368785, + "grad_norm": 3.5626847068687737, + "learning_rate": 9.368311319188293e-07, + "loss": 1.0455, + "step": 5722 + }, + { + "epoch": 0.6881500631275176, + "grad_norm": 1.9058666781883475, + "learning_rate": 9.361714194574515e-07, + "loss": 1.0263, + "step": 5723 + }, + { + "epoch": 0.6882703060181566, + "grad_norm": 0.7618394804913561, + "learning_rate": 9.355118683780228e-07, + "loss": 0.8213, + "step": 5724 + }, + { + "epoch": 0.6883905489087958, + "grad_norm": 2.080993393667093, + "learning_rate": 9.348524787805987e-07, + "loss": 1.0292, + "step": 5725 + }, + { + "epoch": 0.6885107917994349, + "grad_norm": 3.522463374167441, + "learning_rate": 9.341932507652053e-07, + "loss": 1.0883, + "step": 5726 + }, + { + "epoch": 0.6886310346900739, + "grad_norm": 1.7459863548619547, + "learning_rate": 9.335341844318489e-07, + "loss": 1.0229, + "step": 5727 + }, + { + "epoch": 0.6887512775807131, + "grad_norm": 1.8934700736961516, + "learning_rate": 9.328752798805091e-07, + "loss": 0.9705, + "step": 5728 + }, + { + "epoch": 0.6888715204713521, + "grad_norm": 2.430308173775302, + "learning_rate": 9.322165372111399e-07, + "loss": 0.9978, + "step": 5729 + }, + { + "epoch": 0.6889917633619912, + "grad_norm": 2.546702373247662, + "learning_rate": 9.315579565236747e-07, + "loss": 0.9942, + "step": 5730 + }, + { + "epoch": 0.6891120062526304, + "grad_norm": 1.7247313027574545, + "learning_rate": 9.308995379180162e-07, + "loss": 0.9769, + "step": 5731 + }, + { + "epoch": 0.6892322491432694, + "grad_norm": 0.7414312022826887, + "learning_rate": 9.302412814940488e-07, + "loss": 0.8324, + "step": 5732 + }, + { + "epoch": 0.6893524920339085, + "grad_norm": 2.0502108116614295, + "learning_rate": 9.295831873516276e-07, + "loss": 0.9491, + "step": 5733 + }, + { + "epoch": 0.6894727349245476, + "grad_norm": 2.2027536715064637, + "learning_rate": 9.289252555905873e-07, + "loss": 0.9994, + "step": 5734 + }, + { + "epoch": 0.6895929778151867, + "grad_norm": 2.014970066804266, + "learning_rate": 9.282674863107334e-07, + "loss": 0.9952, + "step": 5735 + }, + { + "epoch": 0.6897132207058257, + "grad_norm": 2.089640669893382, + "learning_rate": 9.276098796118488e-07, + "loss": 0.9944, + "step": 5736 + }, + { + "epoch": 0.6898334635964649, + "grad_norm": 1.8126439004534438, + "learning_rate": 9.269524355936938e-07, + "loss": 0.8993, + "step": 5737 + }, + { + "epoch": 0.689953706487104, + "grad_norm": 1.9130409472999115, + "learning_rate": 9.262951543560002e-07, + "loss": 1.0815, + "step": 5738 + }, + { + "epoch": 0.690073949377743, + "grad_norm": 3.950991088532729, + "learning_rate": 9.256380359984795e-07, + "loss": 1.1078, + "step": 5739 + }, + { + "epoch": 0.6901941922683821, + "grad_norm": 2.0057745064050527, + "learning_rate": 9.249810806208139e-07, + "loss": 0.985, + "step": 5740 + }, + { + "epoch": 0.6903144351590212, + "grad_norm": 2.0443936515570713, + "learning_rate": 9.243242883226627e-07, + "loss": 1.0365, + "step": 5741 + }, + { + "epoch": 0.6904346780496603, + "grad_norm": 1.8757765751231898, + "learning_rate": 9.236676592036628e-07, + "loss": 0.9337, + "step": 5742 + }, + { + "epoch": 0.6905549209402994, + "grad_norm": 1.83016234766927, + "learning_rate": 9.230111933634228e-07, + "loss": 0.9739, + "step": 5743 + }, + { + "epoch": 0.6906751638309385, + "grad_norm": 1.551436893415223, + "learning_rate": 9.223548909015288e-07, + "loss": 1.0418, + "step": 5744 + }, + { + "epoch": 0.6907954067215776, + "grad_norm": 2.115797649421688, + "learning_rate": 9.216987519175407e-07, + "loss": 0.959, + "step": 5745 + }, + { + "epoch": 0.6909156496122166, + "grad_norm": 1.8772969481769584, + "learning_rate": 9.210427765109942e-07, + "loss": 0.9224, + "step": 5746 + }, + { + "epoch": 0.6910358925028558, + "grad_norm": 2.228431532908919, + "learning_rate": 9.20386964781402e-07, + "loss": 1.0498, + "step": 5747 + }, + { + "epoch": 0.6911561353934949, + "grad_norm": 2.3263436485433884, + "learning_rate": 9.197313168282472e-07, + "loss": 1.0803, + "step": 5748 + }, + { + "epoch": 0.6912763782841339, + "grad_norm": 2.443625065958806, + "learning_rate": 9.190758327509935e-07, + "loss": 0.9569, + "step": 5749 + }, + { + "epoch": 0.6913966211747731, + "grad_norm": 0.9348265357443394, + "learning_rate": 9.184205126490767e-07, + "loss": 0.8824, + "step": 5750 + }, + { + "epoch": 0.6915168640654121, + "grad_norm": 1.1356491352512499, + "learning_rate": 9.177653566219075e-07, + "loss": 0.834, + "step": 5751 + }, + { + "epoch": 0.6916371069560512, + "grad_norm": 2.4278117602559934, + "learning_rate": 9.171103647688744e-07, + "loss": 1.0002, + "step": 5752 + }, + { + "epoch": 0.6917573498466904, + "grad_norm": 2.571209133715451, + "learning_rate": 9.164555371893367e-07, + "loss": 0.9299, + "step": 5753 + }, + { + "epoch": 0.6918775927373294, + "grad_norm": 1.8608360640828512, + "learning_rate": 9.158008739826333e-07, + "loss": 0.9892, + "step": 5754 + }, + { + "epoch": 0.6919978356279685, + "grad_norm": 2.074650893660523, + "learning_rate": 9.151463752480744e-07, + "loss": 1.103, + "step": 5755 + }, + { + "epoch": 0.6921180785186076, + "grad_norm": 1.4680935630501841, + "learning_rate": 9.144920410849493e-07, + "loss": 1.0393, + "step": 5756 + }, + { + "epoch": 0.6922383214092467, + "grad_norm": 1.8495116001537795, + "learning_rate": 9.138378715925176e-07, + "loss": 1.0364, + "step": 5757 + }, + { + "epoch": 0.6923585642998857, + "grad_norm": 1.7889632151677655, + "learning_rate": 9.131838668700167e-07, + "loss": 1.0455, + "step": 5758 + }, + { + "epoch": 0.6924788071905249, + "grad_norm": 1.8882007030878898, + "learning_rate": 9.125300270166598e-07, + "loss": 1.1039, + "step": 5759 + }, + { + "epoch": 0.692599050081164, + "grad_norm": 1.770367759387581, + "learning_rate": 9.118763521316324e-07, + "loss": 1.094, + "step": 5760 + }, + { + "epoch": 0.692719292971803, + "grad_norm": 1.651586986060687, + "learning_rate": 9.112228423140987e-07, + "loss": 0.9978, + "step": 5761 + }, + { + "epoch": 0.6928395358624422, + "grad_norm": 2.3875989523948125, + "learning_rate": 9.105694976631932e-07, + "loss": 1.1007, + "step": 5762 + }, + { + "epoch": 0.6929597787530812, + "grad_norm": 3.157344212981544, + "learning_rate": 9.099163182780283e-07, + "loss": 0.9644, + "step": 5763 + }, + { + "epoch": 0.6930800216437203, + "grad_norm": 2.669433102521333, + "learning_rate": 9.092633042576916e-07, + "loss": 0.7247, + "step": 5764 + }, + { + "epoch": 0.6932002645343595, + "grad_norm": 1.8283195718104204, + "learning_rate": 9.086104557012446e-07, + "loss": 0.7992, + "step": 5765 + }, + { + "epoch": 0.6933205074249985, + "grad_norm": 1.9640484332155248, + "learning_rate": 9.079577727077239e-07, + "loss": 0.8899, + "step": 5766 + }, + { + "epoch": 0.6934407503156376, + "grad_norm": 2.4212915517602354, + "learning_rate": 9.073052553761404e-07, + "loss": 0.9609, + "step": 5767 + }, + { + "epoch": 0.6935609932062767, + "grad_norm": 1.611127621284748, + "learning_rate": 9.066529038054805e-07, + "loss": 1.0182, + "step": 5768 + }, + { + "epoch": 0.6936812360969158, + "grad_norm": 1.7653554524959632, + "learning_rate": 9.060007180947071e-07, + "loss": 0.9767, + "step": 5769 + }, + { + "epoch": 0.6938014789875548, + "grad_norm": 1.8159177718094475, + "learning_rate": 9.053486983427534e-07, + "loss": 0.9705, + "step": 5770 + }, + { + "epoch": 0.6939217218781939, + "grad_norm": 1.8689960993873547, + "learning_rate": 9.046968446485326e-07, + "loss": 0.9366, + "step": 5771 + }, + { + "epoch": 0.6940419647688331, + "grad_norm": 2.4283893220408164, + "learning_rate": 9.040451571109295e-07, + "loss": 0.94, + "step": 5772 + }, + { + "epoch": 0.6941622076594721, + "grad_norm": 0.8515996662393671, + "learning_rate": 9.033936358288042e-07, + "loss": 0.842, + "step": 5773 + }, + { + "epoch": 0.6942824505501112, + "grad_norm": 1.769687132043007, + "learning_rate": 9.027422809009937e-07, + "loss": 1.0588, + "step": 5774 + }, + { + "epoch": 0.6944026934407503, + "grad_norm": 1.7046791475058178, + "learning_rate": 9.020910924263054e-07, + "loss": 1.0676, + "step": 5775 + }, + { + "epoch": 0.6945229363313894, + "grad_norm": 0.813972898739763, + "learning_rate": 9.014400705035261e-07, + "loss": 0.8194, + "step": 5776 + }, + { + "epoch": 0.6946431792220285, + "grad_norm": 2.6470678157902987, + "learning_rate": 9.00789215231414e-07, + "loss": 1.003, + "step": 5777 + }, + { + "epoch": 0.6947634221126676, + "grad_norm": 1.9995746160169996, + "learning_rate": 9.001385267087056e-07, + "loss": 1.0561, + "step": 5778 + }, + { + "epoch": 0.6948836650033067, + "grad_norm": 1.5977017645660014, + "learning_rate": 8.994880050341072e-07, + "loss": 0.941, + "step": 5779 + }, + { + "epoch": 0.6950039078939457, + "grad_norm": 1.9139579900119328, + "learning_rate": 8.988376503063026e-07, + "loss": 1.0124, + "step": 5780 + }, + { + "epoch": 0.6951241507845849, + "grad_norm": 2.1021877337246226, + "learning_rate": 8.981874626239521e-07, + "loss": 1.0577, + "step": 5781 + }, + { + "epoch": 0.695244393675224, + "grad_norm": 2.0832325974587538, + "learning_rate": 8.975374420856872e-07, + "loss": 1.1189, + "step": 5782 + }, + { + "epoch": 0.695364636565863, + "grad_norm": 2.2785413154973124, + "learning_rate": 8.968875887901157e-07, + "loss": 0.9626, + "step": 5783 + }, + { + "epoch": 0.6954848794565022, + "grad_norm": 2.2073544425972704, + "learning_rate": 8.9623790283582e-07, + "loss": 0.8658, + "step": 5784 + }, + { + "epoch": 0.6956051223471412, + "grad_norm": 2.7453957073387922, + "learning_rate": 8.955883843213561e-07, + "loss": 1.0017, + "step": 5785 + }, + { + "epoch": 0.6957253652377803, + "grad_norm": 2.0000698554237317, + "learning_rate": 8.949390333452569e-07, + "loss": 1.1063, + "step": 5786 + }, + { + "epoch": 0.6958456081284194, + "grad_norm": 1.8409647836834286, + "learning_rate": 8.942898500060279e-07, + "loss": 0.9141, + "step": 5787 + }, + { + "epoch": 0.6959658510190585, + "grad_norm": 2.570870846104221, + "learning_rate": 8.936408344021493e-07, + "loss": 0.9575, + "step": 5788 + }, + { + "epoch": 0.6960860939096976, + "grad_norm": 2.2531310654133967, + "learning_rate": 8.929919866320765e-07, + "loss": 0.9466, + "step": 5789 + }, + { + "epoch": 0.6962063368003367, + "grad_norm": 2.135513216642737, + "learning_rate": 8.923433067942385e-07, + "loss": 1.051, + "step": 5790 + }, + { + "epoch": 0.6963265796909758, + "grad_norm": 1.8446566405405853, + "learning_rate": 8.916947949870417e-07, + "loss": 0.923, + "step": 5791 + }, + { + "epoch": 0.6964468225816148, + "grad_norm": 0.7486053453796272, + "learning_rate": 8.910464513088615e-07, + "loss": 0.8185, + "step": 5792 + }, + { + "epoch": 0.696567065472254, + "grad_norm": 2.0529947596658236, + "learning_rate": 8.903982758580542e-07, + "loss": 1.0244, + "step": 5793 + }, + { + "epoch": 0.696687308362893, + "grad_norm": 2.0743550688638606, + "learning_rate": 8.897502687329457e-07, + "loss": 1.0408, + "step": 5794 + }, + { + "epoch": 0.6968075512535321, + "grad_norm": 5.985945453205367, + "learning_rate": 8.891024300318382e-07, + "loss": 1.0335, + "step": 5795 + }, + { + "epoch": 0.6969277941441713, + "grad_norm": 1.5356595910430078, + "learning_rate": 8.884547598530103e-07, + "loss": 0.9974, + "step": 5796 + }, + { + "epoch": 0.6970480370348103, + "grad_norm": 2.0694388219542224, + "learning_rate": 8.8780725829471e-07, + "loss": 0.9905, + "step": 5797 + }, + { + "epoch": 0.6971682799254494, + "grad_norm": 2.5063429475134757, + "learning_rate": 8.87159925455165e-07, + "loss": 1.0192, + "step": 5798 + }, + { + "epoch": 0.6972885228160886, + "grad_norm": 8.225820538715652, + "learning_rate": 8.865127614325738e-07, + "loss": 0.9687, + "step": 5799 + }, + { + "epoch": 0.6974087657067276, + "grad_norm": 2.4892105930140365, + "learning_rate": 8.85865766325113e-07, + "loss": 0.8994, + "step": 5800 + }, + { + "epoch": 0.6975290085973667, + "grad_norm": 2.4374151948701392, + "learning_rate": 8.852189402309287e-07, + "loss": 0.9612, + "step": 5801 + }, + { + "epoch": 0.6976492514880057, + "grad_norm": 2.7347771921009927, + "learning_rate": 8.845722832481441e-07, + "loss": 0.9756, + "step": 5802 + }, + { + "epoch": 0.6977694943786449, + "grad_norm": 2.0240464638137636, + "learning_rate": 8.83925795474858e-07, + "loss": 1.0156, + "step": 5803 + }, + { + "epoch": 0.6978897372692839, + "grad_norm": 2.353742229268541, + "learning_rate": 8.832794770091414e-07, + "loss": 0.832, + "step": 5804 + }, + { + "epoch": 0.698009980159923, + "grad_norm": 2.400666434425287, + "learning_rate": 8.826333279490401e-07, + "loss": 1.0629, + "step": 5805 + }, + { + "epoch": 0.6981302230505622, + "grad_norm": 2.109024018651466, + "learning_rate": 8.819873483925748e-07, + "loss": 0.9203, + "step": 5806 + }, + { + "epoch": 0.6982504659412012, + "grad_norm": 2.576446634235555, + "learning_rate": 8.81341538437739e-07, + "loss": 0.9803, + "step": 5807 + }, + { + "epoch": 0.6983707088318403, + "grad_norm": 1.6185795049835014, + "learning_rate": 8.80695898182503e-07, + "loss": 0.9149, + "step": 5808 + }, + { + "epoch": 0.6984909517224794, + "grad_norm": 0.8339734400583986, + "learning_rate": 8.800504277248093e-07, + "loss": 0.8887, + "step": 5809 + }, + { + "epoch": 0.6986111946131185, + "grad_norm": 1.847333238686805, + "learning_rate": 8.794051271625753e-07, + "loss": 0.9889, + "step": 5810 + }, + { + "epoch": 0.6987314375037575, + "grad_norm": 2.373098716640523, + "learning_rate": 8.787599965936925e-07, + "loss": 1.068, + "step": 5811 + }, + { + "epoch": 0.6988516803943967, + "grad_norm": 1.717802029057818, + "learning_rate": 8.781150361160261e-07, + "loss": 0.9556, + "step": 5812 + }, + { + "epoch": 0.6989719232850358, + "grad_norm": 1.6745415928556726, + "learning_rate": 8.774702458274181e-07, + "loss": 0.9736, + "step": 5813 + }, + { + "epoch": 0.6990921661756748, + "grad_norm": 3.4806282186741604, + "learning_rate": 8.768256258256799e-07, + "loss": 0.9432, + "step": 5814 + }, + { + "epoch": 0.699212409066314, + "grad_norm": 1.8419456057806216, + "learning_rate": 8.76181176208602e-07, + "loss": 0.9717, + "step": 5815 + }, + { + "epoch": 0.699332651956953, + "grad_norm": 2.9114811895811994, + "learning_rate": 8.755368970739461e-07, + "loss": 0.9727, + "step": 5816 + }, + { + "epoch": 0.6994528948475921, + "grad_norm": 2.257057670948302, + "learning_rate": 8.748927885194479e-07, + "loss": 0.8433, + "step": 5817 + }, + { + "epoch": 0.6995731377382313, + "grad_norm": 0.8124748739612375, + "learning_rate": 8.742488506428209e-07, + "loss": 0.8111, + "step": 5818 + }, + { + "epoch": 0.6996933806288703, + "grad_norm": 1.7527682344257784, + "learning_rate": 8.736050835417466e-07, + "loss": 1.0191, + "step": 5819 + }, + { + "epoch": 0.6998136235195094, + "grad_norm": 2.0518482157469315, + "learning_rate": 8.729614873138862e-07, + "loss": 0.8481, + "step": 5820 + }, + { + "epoch": 0.6999338664101485, + "grad_norm": 1.9710122084411286, + "learning_rate": 8.723180620568716e-07, + "loss": 1.0122, + "step": 5821 + }, + { + "epoch": 0.7000541093007876, + "grad_norm": 2.0064467003913236, + "learning_rate": 8.716748078683116e-07, + "loss": 1.0888, + "step": 5822 + }, + { + "epoch": 0.7001743521914267, + "grad_norm": 2.3120888009358738, + "learning_rate": 8.710317248457855e-07, + "loss": 0.9244, + "step": 5823 + }, + { + "epoch": 0.7002945950820658, + "grad_norm": 1.8664472700795232, + "learning_rate": 8.703888130868482e-07, + "loss": 0.9581, + "step": 5824 + }, + { + "epoch": 0.7004148379727049, + "grad_norm": 2.332734837296446, + "learning_rate": 8.697460726890307e-07, + "loss": 1.0583, + "step": 5825 + }, + { + "epoch": 0.7005350808633439, + "grad_norm": 4.456676258389735, + "learning_rate": 8.691035037498354e-07, + "loss": 1.1443, + "step": 5826 + }, + { + "epoch": 0.7006553237539831, + "grad_norm": 1.7644947451618367, + "learning_rate": 8.684611063667391e-07, + "loss": 0.9592, + "step": 5827 + }, + { + "epoch": 0.7007755666446221, + "grad_norm": 2.175394513535479, + "learning_rate": 8.678188806371935e-07, + "loss": 1.0021, + "step": 5828 + }, + { + "epoch": 0.7008958095352612, + "grad_norm": 1.8533337969173336, + "learning_rate": 8.671768266586228e-07, + "loss": 1.091, + "step": 5829 + }, + { + "epoch": 0.7010160524259004, + "grad_norm": 1.9556112105705417, + "learning_rate": 8.665349445284275e-07, + "loss": 1.019, + "step": 5830 + }, + { + "epoch": 0.7011362953165394, + "grad_norm": 1.5059013311377059, + "learning_rate": 8.658932343439799e-07, + "loss": 1.0471, + "step": 5831 + }, + { + "epoch": 0.7012565382071785, + "grad_norm": 9.740466420874895, + "learning_rate": 8.65251696202627e-07, + "loss": 1.0146, + "step": 5832 + }, + { + "epoch": 0.7013767810978175, + "grad_norm": 2.122172100857793, + "learning_rate": 8.646103302016896e-07, + "loss": 1.1119, + "step": 5833 + }, + { + "epoch": 0.7014970239884567, + "grad_norm": 2.616305894541668, + "learning_rate": 8.639691364384614e-07, + "loss": 1.1218, + "step": 5834 + }, + { + "epoch": 0.7016172668790958, + "grad_norm": 2.301510240384926, + "learning_rate": 8.633281150102136e-07, + "loss": 0.9646, + "step": 5835 + }, + { + "epoch": 0.7017375097697348, + "grad_norm": 2.462706108956031, + "learning_rate": 8.626872660141855e-07, + "loss": 0.9116, + "step": 5836 + }, + { + "epoch": 0.701857752660374, + "grad_norm": 1.9862050906386424, + "learning_rate": 8.620465895475957e-07, + "loss": 0.986, + "step": 5837 + }, + { + "epoch": 0.701977995551013, + "grad_norm": 1.481740038327733, + "learning_rate": 8.614060857076333e-07, + "loss": 0.9902, + "step": 5838 + }, + { + "epoch": 0.7020982384416521, + "grad_norm": 1.9069467740724984, + "learning_rate": 8.60765754591462e-07, + "loss": 0.9839, + "step": 5839 + }, + { + "epoch": 0.7022184813322913, + "grad_norm": 2.0127092904202883, + "learning_rate": 8.601255962962211e-07, + "loss": 0.9678, + "step": 5840 + }, + { + "epoch": 0.7023387242229303, + "grad_norm": 2.8187492455187546, + "learning_rate": 8.594856109190194e-07, + "loss": 0.9606, + "step": 5841 + }, + { + "epoch": 0.7024589671135694, + "grad_norm": 1.7803403974653746, + "learning_rate": 8.588457985569446e-07, + "loss": 0.9277, + "step": 5842 + }, + { + "epoch": 0.7025792100042085, + "grad_norm": 2.018020507526826, + "learning_rate": 8.582061593070542e-07, + "loss": 0.9522, + "step": 5843 + }, + { + "epoch": 0.7026994528948476, + "grad_norm": 2.3506693110084194, + "learning_rate": 8.57566693266383e-07, + "loss": 1.0051, + "step": 5844 + }, + { + "epoch": 0.7028196957854866, + "grad_norm": 2.359449372793033, + "learning_rate": 8.569274005319354e-07, + "loss": 0.9301, + "step": 5845 + }, + { + "epoch": 0.7029399386761258, + "grad_norm": 2.0423913435125334, + "learning_rate": 8.562882812006913e-07, + "loss": 1.033, + "step": 5846 + }, + { + "epoch": 0.7030601815667649, + "grad_norm": 1.903132406386322, + "learning_rate": 8.556493353696066e-07, + "loss": 1.0102, + "step": 5847 + }, + { + "epoch": 0.7031804244574039, + "grad_norm": 3.81859601548527, + "learning_rate": 8.550105631356077e-07, + "loss": 0.9181, + "step": 5848 + }, + { + "epoch": 0.7033006673480431, + "grad_norm": 2.1677289950479697, + "learning_rate": 8.543719645955961e-07, + "loss": 1.0088, + "step": 5849 + }, + { + "epoch": 0.7034209102386821, + "grad_norm": 1.649868673820399, + "learning_rate": 8.537335398464467e-07, + "loss": 0.9835, + "step": 5850 + }, + { + "epoch": 0.7035411531293212, + "grad_norm": 2.8542138719947743, + "learning_rate": 8.53095288985007e-07, + "loss": 1.0884, + "step": 5851 + }, + { + "epoch": 0.7036613960199604, + "grad_norm": 1.6745017976156058, + "learning_rate": 8.524572121081009e-07, + "loss": 1.0629, + "step": 5852 + }, + { + "epoch": 0.7037816389105994, + "grad_norm": 4.1323144969625245, + "learning_rate": 8.518193093125232e-07, + "loss": 0.8604, + "step": 5853 + }, + { + "epoch": 0.7039018818012385, + "grad_norm": 5.148131666810152, + "learning_rate": 8.511815806950436e-07, + "loss": 1.0482, + "step": 5854 + }, + { + "epoch": 0.7040221246918776, + "grad_norm": 1.6924991028415663, + "learning_rate": 8.505440263524044e-07, + "loss": 1.0199, + "step": 5855 + }, + { + "epoch": 0.7041423675825167, + "grad_norm": 2.4868486673092716, + "learning_rate": 8.49906646381322e-07, + "loss": 1.1165, + "step": 5856 + }, + { + "epoch": 0.7042626104731557, + "grad_norm": 1.7371869360917191, + "learning_rate": 8.492694408784884e-07, + "loss": 0.9592, + "step": 5857 + }, + { + "epoch": 0.7043828533637949, + "grad_norm": 2.464384057272381, + "learning_rate": 8.486324099405642e-07, + "loss": 0.8624, + "step": 5858 + }, + { + "epoch": 0.704503096254434, + "grad_norm": 1.6679568859812988, + "learning_rate": 8.479955536641887e-07, + "loss": 0.9844, + "step": 5859 + }, + { + "epoch": 0.704623339145073, + "grad_norm": 2.029999920511479, + "learning_rate": 8.473588721459716e-07, + "loss": 0.8991, + "step": 5860 + }, + { + "epoch": 0.7047435820357122, + "grad_norm": 2.2047978536454207, + "learning_rate": 8.467223654824967e-07, + "loss": 0.9388, + "step": 5861 + }, + { + "epoch": 0.7048638249263512, + "grad_norm": 2.14591160109904, + "learning_rate": 8.460860337703233e-07, + "loss": 0.8672, + "step": 5862 + }, + { + "epoch": 0.7049840678169903, + "grad_norm": 1.99980287772543, + "learning_rate": 8.454498771059797e-07, + "loss": 0.9452, + "step": 5863 + }, + { + "epoch": 0.7051043107076294, + "grad_norm": 2.42960929131438, + "learning_rate": 8.448138955859725e-07, + "loss": 1.0726, + "step": 5864 + }, + { + "epoch": 0.7052245535982685, + "grad_norm": 4.459760894866437, + "learning_rate": 8.44178089306778e-07, + "loss": 1.1374, + "step": 5865 + }, + { + "epoch": 0.7053447964889076, + "grad_norm": 1.8535376206792225, + "learning_rate": 8.4354245836485e-07, + "loss": 1.0109, + "step": 5866 + }, + { + "epoch": 0.7054650393795466, + "grad_norm": 1.733900950488094, + "learning_rate": 8.429070028566108e-07, + "loss": 0.9675, + "step": 5867 + }, + { + "epoch": 0.7055852822701858, + "grad_norm": 1.9084090216451202, + "learning_rate": 8.422717228784586e-07, + "loss": 0.9885, + "step": 5868 + }, + { + "epoch": 0.7057055251608249, + "grad_norm": 1.8480575158057377, + "learning_rate": 8.416366185267663e-07, + "loss": 0.9288, + "step": 5869 + }, + { + "epoch": 0.7058257680514639, + "grad_norm": 1.8404628092835664, + "learning_rate": 8.410016898978778e-07, + "loss": 1.0134, + "step": 5870 + }, + { + "epoch": 0.7059460109421031, + "grad_norm": 2.1393462210972642, + "learning_rate": 8.403669370881115e-07, + "loss": 1.0269, + "step": 5871 + }, + { + "epoch": 0.7060662538327421, + "grad_norm": 1.6624624950796294, + "learning_rate": 8.397323601937587e-07, + "loss": 1.0235, + "step": 5872 + }, + { + "epoch": 0.7061864967233812, + "grad_norm": 2.3039781416279994, + "learning_rate": 8.390979593110838e-07, + "loss": 1.012, + "step": 5873 + }, + { + "epoch": 0.7063067396140204, + "grad_norm": 1.6694550871267395, + "learning_rate": 8.384637345363262e-07, + "loss": 1.0533, + "step": 5874 + }, + { + "epoch": 0.7064269825046594, + "grad_norm": 1.9565602723763784, + "learning_rate": 8.378296859656964e-07, + "loss": 1.003, + "step": 5875 + }, + { + "epoch": 0.7065472253952985, + "grad_norm": 2.444514232660387, + "learning_rate": 8.371958136953792e-07, + "loss": 0.9162, + "step": 5876 + }, + { + "epoch": 0.7066674682859376, + "grad_norm": 2.7970548603895504, + "learning_rate": 8.365621178215326e-07, + "loss": 0.9025, + "step": 5877 + }, + { + "epoch": 0.7067877111765767, + "grad_norm": 2.272378156897649, + "learning_rate": 8.359285984402871e-07, + "loss": 0.9897, + "step": 5878 + }, + { + "epoch": 0.7069079540672157, + "grad_norm": 2.110863167719711, + "learning_rate": 8.352952556477489e-07, + "loss": 0.976, + "step": 5879 + }, + { + "epoch": 0.7070281969578549, + "grad_norm": 1.8597629687316937, + "learning_rate": 8.34662089539993e-07, + "loss": 1.0006, + "step": 5880 + }, + { + "epoch": 0.707148439848494, + "grad_norm": 2.2367305476829866, + "learning_rate": 8.340291002130722e-07, + "loss": 1.0244, + "step": 5881 + }, + { + "epoch": 0.707268682739133, + "grad_norm": 2.323905038899347, + "learning_rate": 8.3339628776301e-07, + "loss": 1.0317, + "step": 5882 + }, + { + "epoch": 0.7073889256297722, + "grad_norm": 2.0149002546512937, + "learning_rate": 8.327636522858033e-07, + "loss": 0.8092, + "step": 5883 + }, + { + "epoch": 0.7075091685204112, + "grad_norm": 1.910802065026036, + "learning_rate": 8.321311938774225e-07, + "loss": 1.0068, + "step": 5884 + }, + { + "epoch": 0.7076294114110503, + "grad_norm": 2.0128012105965425, + "learning_rate": 8.314989126338104e-07, + "loss": 1.0312, + "step": 5885 + }, + { + "epoch": 0.7077496543016895, + "grad_norm": 1.7641846850157832, + "learning_rate": 8.308668086508847e-07, + "loss": 1.08, + "step": 5886 + }, + { + "epoch": 0.7078698971923285, + "grad_norm": 2.058635906678169, + "learning_rate": 8.302348820245342e-07, + "loss": 0.9785, + "step": 5887 + }, + { + "epoch": 0.7079901400829676, + "grad_norm": 2.2267603736088466, + "learning_rate": 8.296031328506232e-07, + "loss": 0.9335, + "step": 5888 + }, + { + "epoch": 0.7081103829736067, + "grad_norm": 1.7432775807750218, + "learning_rate": 8.289715612249857e-07, + "loss": 0.9927, + "step": 5889 + }, + { + "epoch": 0.7082306258642458, + "grad_norm": 2.325677904133682, + "learning_rate": 8.283401672434305e-07, + "loss": 1.0143, + "step": 5890 + }, + { + "epoch": 0.7083508687548848, + "grad_norm": 1.8953132195404523, + "learning_rate": 8.277089510017412e-07, + "loss": 0.9371, + "step": 5891 + }, + { + "epoch": 0.708471111645524, + "grad_norm": 1.7155686499508236, + "learning_rate": 8.270779125956719e-07, + "loss": 1.0599, + "step": 5892 + }, + { + "epoch": 0.7085913545361631, + "grad_norm": 2.12755801004111, + "learning_rate": 8.264470521209505e-07, + "loss": 1.037, + "step": 5893 + }, + { + "epoch": 0.7087115974268021, + "grad_norm": 2.2838177795875283, + "learning_rate": 8.258163696732785e-07, + "loss": 0.9988, + "step": 5894 + }, + { + "epoch": 0.7088318403174413, + "grad_norm": 2.3644036268154487, + "learning_rate": 8.251858653483288e-07, + "loss": 1.0123, + "step": 5895 + }, + { + "epoch": 0.7089520832080803, + "grad_norm": 2.2858225315722795, + "learning_rate": 8.245555392417501e-07, + "loss": 1.0968, + "step": 5896 + }, + { + "epoch": 0.7090723260987194, + "grad_norm": 1.8102571488117976, + "learning_rate": 8.239253914491613e-07, + "loss": 1.022, + "step": 5897 + }, + { + "epoch": 0.7091925689893585, + "grad_norm": 2.260618373701885, + "learning_rate": 8.232954220661556e-07, + "loss": 0.9907, + "step": 5898 + }, + { + "epoch": 0.7093128118799976, + "grad_norm": 2.5500148772759372, + "learning_rate": 8.226656311882989e-07, + "loss": 0.9355, + "step": 5899 + }, + { + "epoch": 0.7094330547706367, + "grad_norm": 2.2095452228242176, + "learning_rate": 8.22036018911129e-07, + "loss": 1.0077, + "step": 5900 + }, + { + "epoch": 0.7095532976612757, + "grad_norm": 2.2577223182276303, + "learning_rate": 8.214065853301599e-07, + "loss": 1.0405, + "step": 5901 + }, + { + "epoch": 0.7096735405519149, + "grad_norm": 0.7960089671142646, + "learning_rate": 8.207773305408734e-07, + "loss": 0.8198, + "step": 5902 + }, + { + "epoch": 0.709793783442554, + "grad_norm": 2.6725045516543617, + "learning_rate": 8.201482546387288e-07, + "loss": 1.0338, + "step": 5903 + }, + { + "epoch": 0.709914026333193, + "grad_norm": 1.854830405201987, + "learning_rate": 8.195193577191553e-07, + "loss": 1.1556, + "step": 5904 + }, + { + "epoch": 0.7100342692238322, + "grad_norm": 1.8731148461826215, + "learning_rate": 8.188906398775579e-07, + "loss": 1.0823, + "step": 5905 + }, + { + "epoch": 0.7101545121144712, + "grad_norm": 1.8590167044986392, + "learning_rate": 8.18262101209311e-07, + "loss": 0.9279, + "step": 5906 + }, + { + "epoch": 0.7102747550051103, + "grad_norm": 2.2047033405343184, + "learning_rate": 8.176337418097626e-07, + "loss": 0.9365, + "step": 5907 + }, + { + "epoch": 0.7103949978957494, + "grad_norm": 2.243724867486061, + "learning_rate": 8.170055617742364e-07, + "loss": 1.0332, + "step": 5908 + }, + { + "epoch": 0.7105152407863885, + "grad_norm": 1.7794975226251861, + "learning_rate": 8.163775611980252e-07, + "loss": 0.9432, + "step": 5909 + }, + { + "epoch": 0.7106354836770276, + "grad_norm": 1.9722171710252625, + "learning_rate": 8.157497401763982e-07, + "loss": 1.0235, + "step": 5910 + }, + { + "epoch": 0.7107557265676667, + "grad_norm": 1.683656483342177, + "learning_rate": 8.151220988045935e-07, + "loss": 1.0167, + "step": 5911 + }, + { + "epoch": 0.7108759694583058, + "grad_norm": 1.7806715444290686, + "learning_rate": 8.144946371778234e-07, + "loss": 1.0635, + "step": 5912 + }, + { + "epoch": 0.7109962123489448, + "grad_norm": 1.811353978870676, + "learning_rate": 8.138673553912751e-07, + "loss": 1.0183, + "step": 5913 + }, + { + "epoch": 0.711116455239584, + "grad_norm": 3.241143115863563, + "learning_rate": 8.132402535401059e-07, + "loss": 0.8064, + "step": 5914 + }, + { + "epoch": 0.711236698130223, + "grad_norm": 4.790085152842732, + "learning_rate": 8.126133317194465e-07, + "loss": 0.9791, + "step": 5915 + }, + { + "epoch": 0.7113569410208621, + "grad_norm": 2.1468874246607017, + "learning_rate": 8.11986590024401e-07, + "loss": 0.9813, + "step": 5916 + }, + { + "epoch": 0.7114771839115013, + "grad_norm": 1.6886625876911436, + "learning_rate": 8.113600285500442e-07, + "loss": 0.9264, + "step": 5917 + }, + { + "epoch": 0.7115974268021403, + "grad_norm": 1.74610556101378, + "learning_rate": 8.107336473914268e-07, + "loss": 0.9777, + "step": 5918 + }, + { + "epoch": 0.7117176696927794, + "grad_norm": 0.7800240625435684, + "learning_rate": 8.101074466435694e-07, + "loss": 0.7953, + "step": 5919 + }, + { + "epoch": 0.7118379125834186, + "grad_norm": 1.7131270938096443, + "learning_rate": 8.094814264014662e-07, + "loss": 0.9173, + "step": 5920 + }, + { + "epoch": 0.7119581554740576, + "grad_norm": 2.095668426254858, + "learning_rate": 8.088555867600844e-07, + "loss": 1.0519, + "step": 5921 + }, + { + "epoch": 0.7120783983646967, + "grad_norm": 1.8374143463724035, + "learning_rate": 8.08229927814362e-07, + "loss": 0.8411, + "step": 5922 + }, + { + "epoch": 0.7121986412553358, + "grad_norm": 1.7613757829443084, + "learning_rate": 8.076044496592134e-07, + "loss": 0.8832, + "step": 5923 + }, + { + "epoch": 0.7123188841459749, + "grad_norm": 2.179769097445018, + "learning_rate": 8.069791523895204e-07, + "loss": 1.0162, + "step": 5924 + }, + { + "epoch": 0.7124391270366139, + "grad_norm": 1.8680648976801046, + "learning_rate": 8.063540361001422e-07, + "loss": 1.0091, + "step": 5925 + }, + { + "epoch": 0.7125593699272531, + "grad_norm": 2.3569915049953747, + "learning_rate": 8.057291008859069e-07, + "loss": 1.0266, + "step": 5926 + }, + { + "epoch": 0.7126796128178922, + "grad_norm": 1.9963453996444094, + "learning_rate": 8.051043468416187e-07, + "loss": 0.921, + "step": 5927 + }, + { + "epoch": 0.7127998557085312, + "grad_norm": 1.9701491334302417, + "learning_rate": 8.044797740620506e-07, + "loss": 1.0587, + "step": 5928 + }, + { + "epoch": 0.7129200985991703, + "grad_norm": 2.13950268359104, + "learning_rate": 8.038553826419494e-07, + "loss": 1.0209, + "step": 5929 + }, + { + "epoch": 0.7130403414898094, + "grad_norm": 1.8264695195271448, + "learning_rate": 8.032311726760364e-07, + "loss": 1.0475, + "step": 5930 + }, + { + "epoch": 0.7131605843804485, + "grad_norm": 1.7313118768644555, + "learning_rate": 8.026071442590022e-07, + "loss": 0.923, + "step": 5931 + }, + { + "epoch": 0.7132808272710875, + "grad_norm": 3.44859354956376, + "learning_rate": 8.019832974855134e-07, + "loss": 1.0483, + "step": 5932 + }, + { + "epoch": 0.7134010701617267, + "grad_norm": 2.2435886378163277, + "learning_rate": 8.013596324502052e-07, + "loss": 1.0619, + "step": 5933 + }, + { + "epoch": 0.7135213130523658, + "grad_norm": 2.4142401669995173, + "learning_rate": 8.007361492476872e-07, + "loss": 1.0245, + "step": 5934 + }, + { + "epoch": 0.7136415559430048, + "grad_norm": 1.542559064726588, + "learning_rate": 8.001128479725426e-07, + "loss": 1.026, + "step": 5935 + }, + { + "epoch": 0.713761798833644, + "grad_norm": 1.6317694585968698, + "learning_rate": 7.994897287193248e-07, + "loss": 1.0448, + "step": 5936 + }, + { + "epoch": 0.713882041724283, + "grad_norm": 2.4107174706816448, + "learning_rate": 7.988667915825605e-07, + "loss": 1.0751, + "step": 5937 + }, + { + "epoch": 0.7140022846149221, + "grad_norm": 2.757032073256731, + "learning_rate": 7.982440366567491e-07, + "loss": 0.9917, + "step": 5938 + }, + { + "epoch": 0.7141225275055613, + "grad_norm": 1.6837992177678476, + "learning_rate": 7.97621464036361e-07, + "loss": 0.9916, + "step": 5939 + }, + { + "epoch": 0.7142427703962003, + "grad_norm": 1.6294705444932394, + "learning_rate": 7.969990738158417e-07, + "loss": 0.92, + "step": 5940 + }, + { + "epoch": 0.7143630132868394, + "grad_norm": 1.9368837668659618, + "learning_rate": 7.963768660896062e-07, + "loss": 1.0882, + "step": 5941 + }, + { + "epoch": 0.7144832561774785, + "grad_norm": 1.9496811679468846, + "learning_rate": 7.957548409520432e-07, + "loss": 1.0637, + "step": 5942 + }, + { + "epoch": 0.7146034990681176, + "grad_norm": 2.278424433533513, + "learning_rate": 7.951329984975135e-07, + "loss": 1.0762, + "step": 5943 + }, + { + "epoch": 0.7147237419587567, + "grad_norm": 0.730735403747115, + "learning_rate": 7.94511338820349e-07, + "loss": 0.7803, + "step": 5944 + }, + { + "epoch": 0.7148439848493958, + "grad_norm": 2.6060087682381803, + "learning_rate": 7.938898620148575e-07, + "loss": 1.0188, + "step": 5945 + }, + { + "epoch": 0.7149642277400349, + "grad_norm": 1.971054423904167, + "learning_rate": 7.932685681753135e-07, + "loss": 0.9478, + "step": 5946 + }, + { + "epoch": 0.7150844706306739, + "grad_norm": 1.7881613381622525, + "learning_rate": 7.92647457395969e-07, + "loss": 0.8615, + "step": 5947 + }, + { + "epoch": 0.7152047135213131, + "grad_norm": 2.351078122968591, + "learning_rate": 7.920265297710444e-07, + "loss": 0.9737, + "step": 5948 + }, + { + "epoch": 0.7153249564119522, + "grad_norm": 2.46237518458959, + "learning_rate": 7.914057853947363e-07, + "loss": 0.9669, + "step": 5949 + }, + { + "epoch": 0.7154451993025912, + "grad_norm": 1.88939431228888, + "learning_rate": 7.907852243612089e-07, + "loss": 0.8647, + "step": 5950 + }, + { + "epoch": 0.7155654421932304, + "grad_norm": 1.8791992847236645, + "learning_rate": 7.901648467646009e-07, + "loss": 0.9613, + "step": 5951 + }, + { + "epoch": 0.7156856850838694, + "grad_norm": 4.086308128804674, + "learning_rate": 7.895446526990244e-07, + "loss": 0.9599, + "step": 5952 + }, + { + "epoch": 0.7158059279745085, + "grad_norm": 1.6148999385719636, + "learning_rate": 7.889246422585609e-07, + "loss": 0.9945, + "step": 5953 + }, + { + "epoch": 0.7159261708651476, + "grad_norm": 1.9372814424360316, + "learning_rate": 7.883048155372675e-07, + "loss": 0.9692, + "step": 5954 + }, + { + "epoch": 0.7160464137557867, + "grad_norm": 2.4077910405613117, + "learning_rate": 7.876851726291698e-07, + "loss": 0.9454, + "step": 5955 + }, + { + "epoch": 0.7161666566464258, + "grad_norm": 1.8699917345323316, + "learning_rate": 7.870657136282666e-07, + "loss": 1.0177, + "step": 5956 + }, + { + "epoch": 0.7162868995370649, + "grad_norm": 1.5658795239326937, + "learning_rate": 7.86446438628531e-07, + "loss": 1.0567, + "step": 5957 + }, + { + "epoch": 0.716407142427704, + "grad_norm": 1.4041687717289904, + "learning_rate": 7.858273477239059e-07, + "loss": 0.806, + "step": 5958 + }, + { + "epoch": 0.716527385318343, + "grad_norm": 1.994741679894125, + "learning_rate": 7.852084410083067e-07, + "loss": 0.9474, + "step": 5959 + }, + { + "epoch": 0.7166476282089821, + "grad_norm": 1.7738260524110125, + "learning_rate": 7.84589718575621e-07, + "loss": 0.8754, + "step": 5960 + }, + { + "epoch": 0.7167678710996213, + "grad_norm": 2.074772015651636, + "learning_rate": 7.83971180519708e-07, + "loss": 0.9279, + "step": 5961 + }, + { + "epoch": 0.7168881139902603, + "grad_norm": 2.618881269638177, + "learning_rate": 7.833528269344008e-07, + "loss": 0.9951, + "step": 5962 + }, + { + "epoch": 0.7170083568808994, + "grad_norm": 3.123387187573538, + "learning_rate": 7.827346579135023e-07, + "loss": 1.01, + "step": 5963 + }, + { + "epoch": 0.7171285997715385, + "grad_norm": 2.0470031232264945, + "learning_rate": 7.821166735507885e-07, + "loss": 1.0666, + "step": 5964 + }, + { + "epoch": 0.7172488426621776, + "grad_norm": 1.9578587044806925, + "learning_rate": 7.81498873940007e-07, + "loss": 0.9253, + "step": 5965 + }, + { + "epoch": 0.7173690855528166, + "grad_norm": 2.2928233839808496, + "learning_rate": 7.808812591748768e-07, + "loss": 1.012, + "step": 5966 + }, + { + "epoch": 0.7174893284434558, + "grad_norm": 2.134357717061478, + "learning_rate": 7.802638293490915e-07, + "loss": 0.8924, + "step": 5967 + }, + { + "epoch": 0.7176095713340949, + "grad_norm": 1.7170548836451638, + "learning_rate": 7.796465845563123e-07, + "loss": 1.0074, + "step": 5968 + }, + { + "epoch": 0.7177298142247339, + "grad_norm": 2.1585001607315606, + "learning_rate": 7.790295248901766e-07, + "loss": 1.0326, + "step": 5969 + }, + { + "epoch": 0.7178500571153731, + "grad_norm": 2.0741962208934224, + "learning_rate": 7.784126504442902e-07, + "loss": 0.8599, + "step": 5970 + }, + { + "epoch": 0.7179703000060121, + "grad_norm": 1.4006347017918213, + "learning_rate": 7.777959613122351e-07, + "loss": 0.9148, + "step": 5971 + }, + { + "epoch": 0.7180905428966512, + "grad_norm": 1.6839504350902303, + "learning_rate": 7.771794575875604e-07, + "loss": 1.0147, + "step": 5972 + }, + { + "epoch": 0.7182107857872904, + "grad_norm": 2.522988483174362, + "learning_rate": 7.765631393637888e-07, + "loss": 1.0095, + "step": 5973 + }, + { + "epoch": 0.7183310286779294, + "grad_norm": 2.8423508775927964, + "learning_rate": 7.75947006734417e-07, + "loss": 0.7232, + "step": 5974 + }, + { + "epoch": 0.7184512715685685, + "grad_norm": 2.168543064237798, + "learning_rate": 7.753310597929101e-07, + "loss": 1.068, + "step": 5975 + }, + { + "epoch": 0.7185715144592076, + "grad_norm": 0.7646448907489782, + "learning_rate": 7.747152986327095e-07, + "loss": 0.789, + "step": 5976 + }, + { + "epoch": 0.7186917573498467, + "grad_norm": 2.1729927754323755, + "learning_rate": 7.740997233472228e-07, + "loss": 0.9127, + "step": 5977 + }, + { + "epoch": 0.7188120002404857, + "grad_norm": 2.028725331568388, + "learning_rate": 7.734843340298329e-07, + "loss": 0.9395, + "step": 5978 + }, + { + "epoch": 0.7189322431311249, + "grad_norm": 2.066162210342998, + "learning_rate": 7.72869130773895e-07, + "loss": 0.9854, + "step": 5979 + }, + { + "epoch": 0.719052486021764, + "grad_norm": 0.8061248016735749, + "learning_rate": 7.722541136727343e-07, + "loss": 0.8314, + "step": 5980 + }, + { + "epoch": 0.719172728912403, + "grad_norm": 1.9752666462393698, + "learning_rate": 7.716392828196483e-07, + "loss": 1.0442, + "step": 5981 + }, + { + "epoch": 0.7192929718030422, + "grad_norm": 3.160421956778122, + "learning_rate": 7.710246383079064e-07, + "loss": 1.0102, + "step": 5982 + }, + { + "epoch": 0.7194132146936812, + "grad_norm": 2.490596155552419, + "learning_rate": 7.704101802307492e-07, + "loss": 1.1555, + "step": 5983 + }, + { + "epoch": 0.7195334575843203, + "grad_norm": 2.331168930648621, + "learning_rate": 7.697959086813912e-07, + "loss": 1.1114, + "step": 5984 + }, + { + "epoch": 0.7196537004749595, + "grad_norm": 2.5334690869080774, + "learning_rate": 7.691818237530145e-07, + "loss": 1.0395, + "step": 5985 + }, + { + "epoch": 0.7197739433655985, + "grad_norm": 2.0133467225814807, + "learning_rate": 7.685679255387774e-07, + "loss": 1.0093, + "step": 5986 + }, + { + "epoch": 0.7198941862562376, + "grad_norm": 2.714191078386239, + "learning_rate": 7.679542141318065e-07, + "loss": 1.0032, + "step": 5987 + }, + { + "epoch": 0.7200144291468767, + "grad_norm": 1.8491297324527547, + "learning_rate": 7.673406896252013e-07, + "loss": 0.9974, + "step": 5988 + }, + { + "epoch": 0.7201346720375158, + "grad_norm": 1.6325872530556282, + "learning_rate": 7.667273521120347e-07, + "loss": 1.0217, + "step": 5989 + }, + { + "epoch": 0.7202549149281549, + "grad_norm": 1.93225846857581, + "learning_rate": 7.661142016853468e-07, + "loss": 1.0336, + "step": 5990 + }, + { + "epoch": 0.7203751578187939, + "grad_norm": 2.1789197218989003, + "learning_rate": 7.655012384381543e-07, + "loss": 0.9875, + "step": 5991 + }, + { + "epoch": 0.7204954007094331, + "grad_norm": 2.140109703246811, + "learning_rate": 7.648884624634415e-07, + "loss": 1.0542, + "step": 5992 + }, + { + "epoch": 0.7206156436000721, + "grad_norm": 1.8450286440189394, + "learning_rate": 7.642758738541683e-07, + "loss": 1.1297, + "step": 5993 + }, + { + "epoch": 0.7207358864907112, + "grad_norm": 0.7728538623049676, + "learning_rate": 7.636634727032621e-07, + "loss": 0.8454, + "step": 5994 + }, + { + "epoch": 0.7208561293813504, + "grad_norm": 2.0704092327047667, + "learning_rate": 7.630512591036231e-07, + "loss": 1.0253, + "step": 5995 + }, + { + "epoch": 0.7209763722719894, + "grad_norm": 2.2247461238423942, + "learning_rate": 7.624392331481255e-07, + "loss": 0.8853, + "step": 5996 + }, + { + "epoch": 0.7210966151626285, + "grad_norm": 0.7647577161396553, + "learning_rate": 7.618273949296115e-07, + "loss": 0.7553, + "step": 5997 + }, + { + "epoch": 0.7212168580532676, + "grad_norm": 1.96033337774581, + "learning_rate": 7.612157445408987e-07, + "loss": 0.9219, + "step": 5998 + }, + { + "epoch": 0.7213371009439067, + "grad_norm": 2.2193672503607593, + "learning_rate": 7.606042820747716e-07, + "loss": 0.9795, + "step": 5999 + }, + { + "epoch": 0.7214573438345457, + "grad_norm": 1.8245474868369478, + "learning_rate": 7.599930076239889e-07, + "loss": 1.0896, + "step": 6000 + }, + { + "epoch": 0.7215775867251849, + "grad_norm": 1.8597934796932092, + "learning_rate": 7.593819212812818e-07, + "loss": 0.9431, + "step": 6001 + }, + { + "epoch": 0.721697829615824, + "grad_norm": 1.7256549504052878, + "learning_rate": 7.587710231393508e-07, + "loss": 0.9535, + "step": 6002 + }, + { + "epoch": 0.721818072506463, + "grad_norm": 2.133585471909289, + "learning_rate": 7.581603132908685e-07, + "loss": 1.0733, + "step": 6003 + }, + { + "epoch": 0.7219383153971022, + "grad_norm": 2.3232232021279224, + "learning_rate": 7.575497918284795e-07, + "loss": 1.0265, + "step": 6004 + }, + { + "epoch": 0.7220585582877412, + "grad_norm": 2.181090419582828, + "learning_rate": 7.569394588447984e-07, + "loss": 0.9846, + "step": 6005 + }, + { + "epoch": 0.7221788011783803, + "grad_norm": 2.5363108587507166, + "learning_rate": 7.563293144324146e-07, + "loss": 1.0215, + "step": 6006 + }, + { + "epoch": 0.7222990440690195, + "grad_norm": 2.026760245199766, + "learning_rate": 7.557193586838834e-07, + "loss": 1.0365, + "step": 6007 + }, + { + "epoch": 0.7224192869596585, + "grad_norm": 2.1616053764852414, + "learning_rate": 7.551095916917371e-07, + "loss": 0.9456, + "step": 6008 + }, + { + "epoch": 0.7225395298502976, + "grad_norm": 2.6165477369295878, + "learning_rate": 7.545000135484758e-07, + "loss": 0.9042, + "step": 6009 + }, + { + "epoch": 0.7226597727409367, + "grad_norm": 1.964388724511872, + "learning_rate": 7.538906243465714e-07, + "loss": 0.8622, + "step": 6010 + }, + { + "epoch": 0.7227800156315758, + "grad_norm": 2.0720749748465903, + "learning_rate": 7.5328142417847e-07, + "loss": 1.0233, + "step": 6011 + }, + { + "epoch": 0.7229002585222148, + "grad_norm": 2.9163240140637465, + "learning_rate": 7.526724131365838e-07, + "loss": 0.9297, + "step": 6012 + }, + { + "epoch": 0.723020501412854, + "grad_norm": 1.7393855669671778, + "learning_rate": 7.520635913133017e-07, + "loss": 0.9387, + "step": 6013 + }, + { + "epoch": 0.7231407443034931, + "grad_norm": 1.8473181384941713, + "learning_rate": 7.514549588009798e-07, + "loss": 1.0624, + "step": 6014 + }, + { + "epoch": 0.7232609871941321, + "grad_norm": 1.9315542851699063, + "learning_rate": 7.508465156919492e-07, + "loss": 0.94, + "step": 6015 + }, + { + "epoch": 0.7233812300847713, + "grad_norm": 2.789836017616087, + "learning_rate": 7.502382620785083e-07, + "loss": 0.8478, + "step": 6016 + }, + { + "epoch": 0.7235014729754103, + "grad_norm": 0.8146086253917036, + "learning_rate": 7.496301980529289e-07, + "loss": 0.8628, + "step": 6017 + }, + { + "epoch": 0.7236217158660494, + "grad_norm": 2.208859387006095, + "learning_rate": 7.490223237074547e-07, + "loss": 0.9812, + "step": 6018 + }, + { + "epoch": 0.7237419587566886, + "grad_norm": 1.9884981587923967, + "learning_rate": 7.484146391342989e-07, + "loss": 0.8931, + "step": 6019 + }, + { + "epoch": 0.7238622016473276, + "grad_norm": 2.8109433422695855, + "learning_rate": 7.478071444256484e-07, + "loss": 0.806, + "step": 6020 + }, + { + "epoch": 0.7239824445379667, + "grad_norm": 1.8189742117254888, + "learning_rate": 7.471998396736579e-07, + "loss": 1.0287, + "step": 6021 + }, + { + "epoch": 0.7241026874286057, + "grad_norm": 1.859482577762086, + "learning_rate": 7.465927249704549e-07, + "loss": 0.9986, + "step": 6022 + }, + { + "epoch": 0.7242229303192449, + "grad_norm": 2.0151399723781886, + "learning_rate": 7.459858004081398e-07, + "loss": 1.0099, + "step": 6023 + }, + { + "epoch": 0.724343173209884, + "grad_norm": 0.8700221359791132, + "learning_rate": 7.453790660787815e-07, + "loss": 0.8178, + "step": 6024 + }, + { + "epoch": 0.724463416100523, + "grad_norm": 2.02172213275714, + "learning_rate": 7.447725220744214e-07, + "loss": 0.8691, + "step": 6025 + }, + { + "epoch": 0.7245836589911622, + "grad_norm": 2.3639369086610333, + "learning_rate": 7.441661684870717e-07, + "loss": 1.0081, + "step": 6026 + }, + { + "epoch": 0.7247039018818012, + "grad_norm": 1.6290114846758479, + "learning_rate": 7.435600054087152e-07, + "loss": 1.0528, + "step": 6027 + }, + { + "epoch": 0.7248241447724403, + "grad_norm": 2.1363407895620794, + "learning_rate": 7.42954032931308e-07, + "loss": 0.9792, + "step": 6028 + }, + { + "epoch": 0.7249443876630794, + "grad_norm": 3.925604274423564, + "learning_rate": 7.423482511467733e-07, + "loss": 0.9842, + "step": 6029 + }, + { + "epoch": 0.7250646305537185, + "grad_norm": 2.4982516851267746, + "learning_rate": 7.417426601470099e-07, + "loss": 0.884, + "step": 6030 + }, + { + "epoch": 0.7251848734443576, + "grad_norm": 2.1226345809487563, + "learning_rate": 7.411372600238841e-07, + "loss": 1.0256, + "step": 6031 + }, + { + "epoch": 0.7253051163349967, + "grad_norm": 2.5963775929599704, + "learning_rate": 7.405320508692346e-07, + "loss": 0.9753, + "step": 6032 + }, + { + "epoch": 0.7254253592256358, + "grad_norm": 1.903548029845507, + "learning_rate": 7.399270327748727e-07, + "loss": 0.9904, + "step": 6033 + }, + { + "epoch": 0.7255456021162748, + "grad_norm": 1.8759846327219374, + "learning_rate": 7.39322205832577e-07, + "loss": 0.9775, + "step": 6034 + }, + { + "epoch": 0.725665845006914, + "grad_norm": 1.9193796224485662, + "learning_rate": 7.387175701341009e-07, + "loss": 1.048, + "step": 6035 + }, + { + "epoch": 0.7257860878975531, + "grad_norm": 6.7922532778043365, + "learning_rate": 7.381131257711659e-07, + "loss": 0.9575, + "step": 6036 + }, + { + "epoch": 0.7259063307881921, + "grad_norm": 1.8672298921377553, + "learning_rate": 7.375088728354677e-07, + "loss": 1.0729, + "step": 6037 + }, + { + "epoch": 0.7260265736788313, + "grad_norm": 1.6310975043681795, + "learning_rate": 7.369048114186691e-07, + "loss": 0.9091, + "step": 6038 + }, + { + "epoch": 0.7261468165694703, + "grad_norm": 1.7589825614294656, + "learning_rate": 7.363009416124055e-07, + "loss": 1.068, + "step": 6039 + }, + { + "epoch": 0.7262670594601094, + "grad_norm": 2.651455849979337, + "learning_rate": 7.356972635082852e-07, + "loss": 0.8633, + "step": 6040 + }, + { + "epoch": 0.7263873023507486, + "grad_norm": 2.0157804207430132, + "learning_rate": 7.35093777197884e-07, + "loss": 0.9901, + "step": 6041 + }, + { + "epoch": 0.7265075452413876, + "grad_norm": 2.4999845504283362, + "learning_rate": 7.344904827727525e-07, + "loss": 1.0913, + "step": 6042 + }, + { + "epoch": 0.7266277881320267, + "grad_norm": 2.8734893146964726, + "learning_rate": 7.338873803244076e-07, + "loss": 0.9723, + "step": 6043 + }, + { + "epoch": 0.7267480310226658, + "grad_norm": 1.815216922039914, + "learning_rate": 7.332844699443401e-07, + "loss": 1.0454, + "step": 6044 + }, + { + "epoch": 0.7268682739133049, + "grad_norm": 1.8604273022229838, + "learning_rate": 7.326817517240121e-07, + "loss": 0.9915, + "step": 6045 + }, + { + "epoch": 0.7269885168039439, + "grad_norm": 1.7869132512772778, + "learning_rate": 7.320792257548545e-07, + "loss": 1.0722, + "step": 6046 + }, + { + "epoch": 0.7271087596945831, + "grad_norm": 2.2640778619364856, + "learning_rate": 7.314768921282704e-07, + "loss": 1.0016, + "step": 6047 + }, + { + "epoch": 0.7272290025852222, + "grad_norm": 3.492114994397269, + "learning_rate": 7.30874750935633e-07, + "loss": 0.9593, + "step": 6048 + }, + { + "epoch": 0.7273492454758612, + "grad_norm": 1.8636417320114935, + "learning_rate": 7.30272802268286e-07, + "loss": 1.0293, + "step": 6049 + }, + { + "epoch": 0.7274694883665004, + "grad_norm": 1.774786818153997, + "learning_rate": 7.29671046217547e-07, + "loss": 0.9965, + "step": 6050 + }, + { + "epoch": 0.7275897312571394, + "grad_norm": 1.894449951944459, + "learning_rate": 7.290694828746988e-07, + "loss": 1.0547, + "step": 6051 + }, + { + "epoch": 0.7277099741477785, + "grad_norm": 1.7945630002233497, + "learning_rate": 7.284681123310004e-07, + "loss": 1.096, + "step": 6052 + }, + { + "epoch": 0.7278302170384175, + "grad_norm": 1.7608147646924937, + "learning_rate": 7.27866934677678e-07, + "loss": 1.0346, + "step": 6053 + }, + { + "epoch": 0.7279504599290567, + "grad_norm": 1.67030053558204, + "learning_rate": 7.272659500059297e-07, + "loss": 1.0206, + "step": 6054 + }, + { + "epoch": 0.7280707028196958, + "grad_norm": 2.2726719147269203, + "learning_rate": 7.266651584069264e-07, + "loss": 1.0371, + "step": 6055 + }, + { + "epoch": 0.7281909457103348, + "grad_norm": 1.7142374622275343, + "learning_rate": 7.260645599718045e-07, + "loss": 0.8051, + "step": 6056 + }, + { + "epoch": 0.728311188600974, + "grad_norm": 2.72822314738011, + "learning_rate": 7.254641547916767e-07, + "loss": 0.9077, + "step": 6057 + }, + { + "epoch": 0.728431431491613, + "grad_norm": 1.7247772653495084, + "learning_rate": 7.248639429576226e-07, + "loss": 0.9346, + "step": 6058 + }, + { + "epoch": 0.7285516743822521, + "grad_norm": 1.9728281078449352, + "learning_rate": 7.242639245606959e-07, + "loss": 0.9593, + "step": 6059 + }, + { + "epoch": 0.7286719172728913, + "grad_norm": 1.7559326971273188, + "learning_rate": 7.236640996919168e-07, + "loss": 1.0595, + "step": 6060 + }, + { + "epoch": 0.7287921601635303, + "grad_norm": 1.59585350196517, + "learning_rate": 7.230644684422782e-07, + "loss": 0.9383, + "step": 6061 + }, + { + "epoch": 0.7289124030541694, + "grad_norm": 1.695939049327305, + "learning_rate": 7.224650309027451e-07, + "loss": 1.0506, + "step": 6062 + }, + { + "epoch": 0.7290326459448085, + "grad_norm": 1.9345655060910132, + "learning_rate": 7.218657871642506e-07, + "loss": 0.925, + "step": 6063 + }, + { + "epoch": 0.7291528888354476, + "grad_norm": 2.291619236773177, + "learning_rate": 7.212667373177012e-07, + "loss": 0.8625, + "step": 6064 + }, + { + "epoch": 0.7292731317260867, + "grad_norm": 1.8641469931498702, + "learning_rate": 7.206678814539704e-07, + "loss": 0.9883, + "step": 6065 + }, + { + "epoch": 0.7293933746167258, + "grad_norm": 2.0160744336026317, + "learning_rate": 7.20069219663904e-07, + "loss": 0.9659, + "step": 6066 + }, + { + "epoch": 0.7295136175073649, + "grad_norm": 2.459969660253987, + "learning_rate": 7.1947075203832e-07, + "loss": 1.0333, + "step": 6067 + }, + { + "epoch": 0.7296338603980039, + "grad_norm": 0.8736750243931041, + "learning_rate": 7.188724786680049e-07, + "loss": 0.8389, + "step": 6068 + }, + { + "epoch": 0.7297541032886431, + "grad_norm": 1.5337443374491952, + "learning_rate": 7.182743996437162e-07, + "loss": 0.9941, + "step": 6069 + }, + { + "epoch": 0.7298743461792822, + "grad_norm": 2.1947445389283935, + "learning_rate": 7.176765150561819e-07, + "loss": 0.9274, + "step": 6070 + }, + { + "epoch": 0.7299945890699212, + "grad_norm": 2.107951270103013, + "learning_rate": 7.170788249961002e-07, + "loss": 1.0363, + "step": 6071 + }, + { + "epoch": 0.7301148319605604, + "grad_norm": 2.124646325692195, + "learning_rate": 7.164813295541418e-07, + "loss": 1.1162, + "step": 6072 + }, + { + "epoch": 0.7302350748511994, + "grad_norm": 1.8397837505559762, + "learning_rate": 7.15884028820944e-07, + "loss": 0.939, + "step": 6073 + }, + { + "epoch": 0.7303553177418385, + "grad_norm": 2.9828701519277905, + "learning_rate": 7.152869228871185e-07, + "loss": 0.8397, + "step": 6074 + }, + { + "epoch": 0.7304755606324776, + "grad_norm": 1.9290421986094843, + "learning_rate": 7.146900118432457e-07, + "loss": 0.9615, + "step": 6075 + }, + { + "epoch": 0.7305958035231167, + "grad_norm": 1.7039282068432131, + "learning_rate": 7.140932957798753e-07, + "loss": 1.0936, + "step": 6076 + }, + { + "epoch": 0.7307160464137558, + "grad_norm": 1.9190610428484522, + "learning_rate": 7.134967747875309e-07, + "loss": 0.951, + "step": 6077 + }, + { + "epoch": 0.7308362893043949, + "grad_norm": 1.9931648400177624, + "learning_rate": 7.129004489567014e-07, + "loss": 1.0544, + "step": 6078 + }, + { + "epoch": 0.730956532195034, + "grad_norm": 2.243963196382951, + "learning_rate": 7.123043183778512e-07, + "loss": 1.0163, + "step": 6079 + }, + { + "epoch": 0.731076775085673, + "grad_norm": 1.8442923184680033, + "learning_rate": 7.117083831414114e-07, + "loss": 0.8918, + "step": 6080 + }, + { + "epoch": 0.7311970179763122, + "grad_norm": 1.88604503157527, + "learning_rate": 7.11112643337787e-07, + "loss": 0.9309, + "step": 6081 + }, + { + "epoch": 0.7313172608669513, + "grad_norm": 3.625276620768049, + "learning_rate": 7.10517099057349e-07, + "loss": 1.0045, + "step": 6082 + }, + { + "epoch": 0.7314375037575903, + "grad_norm": 2.298834866341273, + "learning_rate": 7.099217503904411e-07, + "loss": 0.8452, + "step": 6083 + }, + { + "epoch": 0.7315577466482295, + "grad_norm": 2.5993809990195675, + "learning_rate": 7.093265974273788e-07, + "loss": 1.1351, + "step": 6084 + }, + { + "epoch": 0.7316779895388685, + "grad_norm": 2.200994340562629, + "learning_rate": 7.087316402584447e-07, + "loss": 0.9569, + "step": 6085 + }, + { + "epoch": 0.7317982324295076, + "grad_norm": 1.7995996135879988, + "learning_rate": 7.081368789738953e-07, + "loss": 1.102, + "step": 6086 + }, + { + "epoch": 0.7319184753201466, + "grad_norm": 2.293106204954255, + "learning_rate": 7.075423136639537e-07, + "loss": 1.0151, + "step": 6087 + }, + { + "epoch": 0.7320387182107858, + "grad_norm": 1.8834108533630993, + "learning_rate": 7.069479444188149e-07, + "loss": 0.9824, + "step": 6088 + }, + { + "epoch": 0.7321589611014249, + "grad_norm": 1.704824387240503, + "learning_rate": 7.063537713286453e-07, + "loss": 1.0566, + "step": 6089 + }, + { + "epoch": 0.7322792039920639, + "grad_norm": 1.9004523491710406, + "learning_rate": 7.057597944835803e-07, + "loss": 1.0419, + "step": 6090 + }, + { + "epoch": 0.7323994468827031, + "grad_norm": 1.7198240912107443, + "learning_rate": 7.051660139737253e-07, + "loss": 0.9862, + "step": 6091 + }, + { + "epoch": 0.7325196897733421, + "grad_norm": 1.9977207309190883, + "learning_rate": 7.045724298891565e-07, + "loss": 1.0025, + "step": 6092 + }, + { + "epoch": 0.7326399326639812, + "grad_norm": 7.690302144279671, + "learning_rate": 7.039790423199192e-07, + "loss": 0.9312, + "step": 6093 + }, + { + "epoch": 0.7327601755546204, + "grad_norm": 3.2123059700284724, + "learning_rate": 7.033858513560322e-07, + "loss": 1.0153, + "step": 6094 + }, + { + "epoch": 0.7328804184452594, + "grad_norm": 2.6185928443239286, + "learning_rate": 7.027928570874794e-07, + "loss": 1.0019, + "step": 6095 + }, + { + "epoch": 0.7330006613358985, + "grad_norm": 2.1559031110731106, + "learning_rate": 7.022000596042194e-07, + "loss": 1.0952, + "step": 6096 + }, + { + "epoch": 0.7331209042265376, + "grad_norm": 2.1509573291073503, + "learning_rate": 7.016074589961784e-07, + "loss": 1.0559, + "step": 6097 + }, + { + "epoch": 0.7332411471171767, + "grad_norm": 1.7128547829996135, + "learning_rate": 7.01015055353253e-07, + "loss": 0.9051, + "step": 6098 + }, + { + "epoch": 0.7333613900078157, + "grad_norm": 2.1021135593304185, + "learning_rate": 7.004228487653123e-07, + "loss": 1.0183, + "step": 6099 + }, + { + "epoch": 0.7334816328984549, + "grad_norm": 2.014842864883542, + "learning_rate": 6.998308393221906e-07, + "loss": 1.0225, + "step": 6100 + }, + { + "epoch": 0.733601875789094, + "grad_norm": 2.720369492426256, + "learning_rate": 6.992390271136977e-07, + "loss": 0.9512, + "step": 6101 + }, + { + "epoch": 0.733722118679733, + "grad_norm": 1.8719062076571846, + "learning_rate": 6.986474122296094e-07, + "loss": 1.0954, + "step": 6102 + }, + { + "epoch": 0.7338423615703722, + "grad_norm": 1.8845154591303552, + "learning_rate": 6.980559947596751e-07, + "loss": 0.9598, + "step": 6103 + }, + { + "epoch": 0.7339626044610112, + "grad_norm": 2.812011845928652, + "learning_rate": 6.974647747936109e-07, + "loss": 0.9947, + "step": 6104 + }, + { + "epoch": 0.7340828473516503, + "grad_norm": 1.9914572181330064, + "learning_rate": 6.968737524211039e-07, + "loss": 1.0611, + "step": 6105 + }, + { + "epoch": 0.7342030902422895, + "grad_norm": 2.218738824520711, + "learning_rate": 6.962829277318132e-07, + "loss": 1.0411, + "step": 6106 + }, + { + "epoch": 0.7343233331329285, + "grad_norm": 4.252298967475013, + "learning_rate": 6.956923008153652e-07, + "loss": 1.0728, + "step": 6107 + }, + { + "epoch": 0.7344435760235676, + "grad_norm": 2.2644574544173293, + "learning_rate": 6.951018717613593e-07, + "loss": 1.0809, + "step": 6108 + }, + { + "epoch": 0.7345638189142067, + "grad_norm": 2.1320745847441898, + "learning_rate": 6.945116406593614e-07, + "loss": 1.02, + "step": 6109 + }, + { + "epoch": 0.7346840618048458, + "grad_norm": 2.6748860664703855, + "learning_rate": 6.939216075989089e-07, + "loss": 0.9769, + "step": 6110 + }, + { + "epoch": 0.7348043046954849, + "grad_norm": 2.054077405969789, + "learning_rate": 6.933317726695109e-07, + "loss": 0.8997, + "step": 6111 + }, + { + "epoch": 0.734924547586124, + "grad_norm": 3.7653852837583623, + "learning_rate": 6.92742135960644e-07, + "loss": 1.0349, + "step": 6112 + }, + { + "epoch": 0.7350447904767631, + "grad_norm": 0.823448544451716, + "learning_rate": 6.921526975617556e-07, + "loss": 0.7945, + "step": 6113 + }, + { + "epoch": 0.7351650333674021, + "grad_norm": 3.4977897749774103, + "learning_rate": 6.915634575622631e-07, + "loss": 0.9926, + "step": 6114 + }, + { + "epoch": 0.7352852762580413, + "grad_norm": 1.769579754095168, + "learning_rate": 6.909744160515532e-07, + "loss": 0.9458, + "step": 6115 + }, + { + "epoch": 0.7354055191486804, + "grad_norm": 1.7565681630891439, + "learning_rate": 6.903855731189849e-07, + "loss": 0.9306, + "step": 6116 + }, + { + "epoch": 0.7355257620393194, + "grad_norm": 2.3512662276377765, + "learning_rate": 6.897969288538825e-07, + "loss": 1.0559, + "step": 6117 + }, + { + "epoch": 0.7356460049299585, + "grad_norm": 1.9918939112330198, + "learning_rate": 6.892084833455452e-07, + "loss": 1.0489, + "step": 6118 + }, + { + "epoch": 0.7357662478205976, + "grad_norm": 1.5319165705779887, + "learning_rate": 6.886202366832384e-07, + "loss": 1.0776, + "step": 6119 + }, + { + "epoch": 0.7358864907112367, + "grad_norm": 1.7457003224884502, + "learning_rate": 6.880321889561987e-07, + "loss": 0.9693, + "step": 6120 + }, + { + "epoch": 0.7360067336018757, + "grad_norm": 2.014039828318624, + "learning_rate": 6.874443402536338e-07, + "loss": 0.8856, + "step": 6121 + }, + { + "epoch": 0.7361269764925149, + "grad_norm": 1.6230913837816408, + "learning_rate": 6.868566906647177e-07, + "loss": 1.0414, + "step": 6122 + }, + { + "epoch": 0.736247219383154, + "grad_norm": 1.8448842972242605, + "learning_rate": 6.862692402785984e-07, + "loss": 1.0695, + "step": 6123 + }, + { + "epoch": 0.736367462273793, + "grad_norm": 0.7034047100232127, + "learning_rate": 6.856819891843899e-07, + "loss": 0.7335, + "step": 6124 + }, + { + "epoch": 0.7364877051644322, + "grad_norm": 2.2121065641261057, + "learning_rate": 6.8509493747118e-07, + "loss": 0.9586, + "step": 6125 + }, + { + "epoch": 0.7366079480550712, + "grad_norm": 2.234960626003594, + "learning_rate": 6.845080852280221e-07, + "loss": 1.1182, + "step": 6126 + }, + { + "epoch": 0.7367281909457103, + "grad_norm": 2.2512442009748956, + "learning_rate": 6.839214325439409e-07, + "loss": 0.9788, + "step": 6127 + }, + { + "epoch": 0.7368484338363495, + "grad_norm": 1.7161541843548789, + "learning_rate": 6.833349795079327e-07, + "loss": 0.9541, + "step": 6128 + }, + { + "epoch": 0.7369686767269885, + "grad_norm": 1.6911836614236373, + "learning_rate": 6.827487262089613e-07, + "loss": 0.921, + "step": 6129 + }, + { + "epoch": 0.7370889196176276, + "grad_norm": 0.9278739422670461, + "learning_rate": 6.821626727359606e-07, + "loss": 0.8044, + "step": 6130 + }, + { + "epoch": 0.7372091625082667, + "grad_norm": 2.480790052855428, + "learning_rate": 6.815768191778348e-07, + "loss": 1.0048, + "step": 6131 + }, + { + "epoch": 0.7373294053989058, + "grad_norm": 1.717453693170982, + "learning_rate": 6.809911656234569e-07, + "loss": 0.965, + "step": 6132 + }, + { + "epoch": 0.7374496482895448, + "grad_norm": 2.0593280098030107, + "learning_rate": 6.804057121616707e-07, + "loss": 1.0183, + "step": 6133 + }, + { + "epoch": 0.737569891180184, + "grad_norm": 1.851110990818646, + "learning_rate": 6.798204588812888e-07, + "loss": 0.9559, + "step": 6134 + }, + { + "epoch": 0.7376901340708231, + "grad_norm": 1.9747646432988768, + "learning_rate": 6.792354058710937e-07, + "loss": 0.9928, + "step": 6135 + }, + { + "epoch": 0.7378103769614621, + "grad_norm": 2.243208915052779, + "learning_rate": 6.786505532198374e-07, + "loss": 0.8888, + "step": 6136 + }, + { + "epoch": 0.7379306198521013, + "grad_norm": 2.6665826327275717, + "learning_rate": 6.780659010162411e-07, + "loss": 1.0899, + "step": 6137 + }, + { + "epoch": 0.7380508627427403, + "grad_norm": 1.6865480351600852, + "learning_rate": 6.774814493489975e-07, + "loss": 1.0693, + "step": 6138 + }, + { + "epoch": 0.7381711056333794, + "grad_norm": 1.7888537955675707, + "learning_rate": 6.768971983067655e-07, + "loss": 0.8983, + "step": 6139 + }, + { + "epoch": 0.7382913485240186, + "grad_norm": 1.0154108261819101, + "learning_rate": 6.763131479781772e-07, + "loss": 0.9153, + "step": 6140 + }, + { + "epoch": 0.7384115914146576, + "grad_norm": 1.9430085292006591, + "learning_rate": 6.757292984518316e-07, + "loss": 0.9968, + "step": 6141 + }, + { + "epoch": 0.7385318343052967, + "grad_norm": 0.7585973706144288, + "learning_rate": 6.751456498162981e-07, + "loss": 0.8021, + "step": 6142 + }, + { + "epoch": 0.7386520771959358, + "grad_norm": 1.9610609038532756, + "learning_rate": 6.745622021601174e-07, + "loss": 1.0911, + "step": 6143 + }, + { + "epoch": 0.7387723200865749, + "grad_norm": 1.7863040631323477, + "learning_rate": 6.739789555717954e-07, + "loss": 0.9347, + "step": 6144 + }, + { + "epoch": 0.738892562977214, + "grad_norm": 2.074516433073334, + "learning_rate": 6.733959101398124e-07, + "loss": 1.0108, + "step": 6145 + }, + { + "epoch": 0.7390128058678531, + "grad_norm": 2.09956088697645, + "learning_rate": 6.728130659526143e-07, + "loss": 1.0524, + "step": 6146 + }, + { + "epoch": 0.7391330487584922, + "grad_norm": 2.385044140997058, + "learning_rate": 6.7223042309862e-07, + "loss": 0.9468, + "step": 6147 + }, + { + "epoch": 0.7392532916491312, + "grad_norm": 2.0197221381693513, + "learning_rate": 6.716479816662144e-07, + "loss": 0.968, + "step": 6148 + }, + { + "epoch": 0.7393735345397703, + "grad_norm": 2.0313654940122596, + "learning_rate": 6.710657417437531e-07, + "loss": 0.9694, + "step": 6149 + }, + { + "epoch": 0.7394937774304094, + "grad_norm": 2.261585607216926, + "learning_rate": 6.704837034195628e-07, + "loss": 1.0334, + "step": 6150 + }, + { + "epoch": 0.7396140203210485, + "grad_norm": 1.7929692154096277, + "learning_rate": 6.699018667819376e-07, + "loss": 1.0827, + "step": 6151 + }, + { + "epoch": 0.7397342632116876, + "grad_norm": 1.6619547358476245, + "learning_rate": 6.693202319191415e-07, + "loss": 0.9681, + "step": 6152 + }, + { + "epoch": 0.7398545061023267, + "grad_norm": 1.9860064190686875, + "learning_rate": 6.687387989194084e-07, + "loss": 0.9819, + "step": 6153 + }, + { + "epoch": 0.7399747489929658, + "grad_norm": 2.9304585573877993, + "learning_rate": 6.681575678709404e-07, + "loss": 1.0308, + "step": 6154 + }, + { + "epoch": 0.7400949918836048, + "grad_norm": 2.536947831083237, + "learning_rate": 6.67576538861911e-07, + "loss": 0.9447, + "step": 6155 + }, + { + "epoch": 0.740215234774244, + "grad_norm": 1.620516239846413, + "learning_rate": 6.669957119804612e-07, + "loss": 1.06, + "step": 6156 + }, + { + "epoch": 0.7403354776648831, + "grad_norm": 2.933774980618291, + "learning_rate": 6.66415087314702e-07, + "loss": 0.9595, + "step": 6157 + }, + { + "epoch": 0.7404557205555221, + "grad_norm": 2.38391788045528, + "learning_rate": 6.65834664952714e-07, + "loss": 0.9718, + "step": 6158 + }, + { + "epoch": 0.7405759634461613, + "grad_norm": 1.5629604424115977, + "learning_rate": 6.652544449825457e-07, + "loss": 0.9969, + "step": 6159 + }, + { + "epoch": 0.7406962063368003, + "grad_norm": 2.08694433708884, + "learning_rate": 6.646744274922182e-07, + "loss": 1.0022, + "step": 6160 + }, + { + "epoch": 0.7408164492274394, + "grad_norm": 4.236548735034942, + "learning_rate": 6.640946125697171e-07, + "loss": 0.9901, + "step": 6161 + }, + { + "epoch": 0.7409366921180786, + "grad_norm": 2.0320057563399834, + "learning_rate": 6.635150003030017e-07, + "loss": 0.9969, + "step": 6162 + }, + { + "epoch": 0.7410569350087176, + "grad_norm": 2.4593945707247746, + "learning_rate": 6.629355907799981e-07, + "loss": 1.094, + "step": 6163 + }, + { + "epoch": 0.7411771778993567, + "grad_norm": 1.8381150338720866, + "learning_rate": 6.623563840886015e-07, + "loss": 0.9295, + "step": 6164 + }, + { + "epoch": 0.7412974207899958, + "grad_norm": 1.6737460780649767, + "learning_rate": 6.617773803166795e-07, + "loss": 0.9314, + "step": 6165 + }, + { + "epoch": 0.7414176636806349, + "grad_norm": 2.5070360354255707, + "learning_rate": 6.611985795520634e-07, + "loss": 1.051, + "step": 6166 + }, + { + "epoch": 0.7415379065712739, + "grad_norm": 2.101305297087647, + "learning_rate": 6.606199818825588e-07, + "loss": 1.0204, + "step": 6167 + }, + { + "epoch": 0.7416581494619131, + "grad_norm": 2.121447791801015, + "learning_rate": 6.600415873959377e-07, + "loss": 1.0499, + "step": 6168 + }, + { + "epoch": 0.7417783923525522, + "grad_norm": 1.9558272322353822, + "learning_rate": 6.594633961799437e-07, + "loss": 0.8848, + "step": 6169 + }, + { + "epoch": 0.7418986352431912, + "grad_norm": 1.9319782945507527, + "learning_rate": 6.588854083222857e-07, + "loss": 1.0555, + "step": 6170 + }, + { + "epoch": 0.7420188781338304, + "grad_norm": 2.107836239782164, + "learning_rate": 6.583076239106444e-07, + "loss": 1.0464, + "step": 6171 + }, + { + "epoch": 0.7421391210244694, + "grad_norm": 2.5259162373552226, + "learning_rate": 6.577300430326707e-07, + "loss": 0.9952, + "step": 6172 + }, + { + "epoch": 0.7422593639151085, + "grad_norm": 2.5718362575871434, + "learning_rate": 6.571526657759821e-07, + "loss": 0.9588, + "step": 6173 + }, + { + "epoch": 0.7423796068057477, + "grad_norm": 1.7777607522917465, + "learning_rate": 6.565754922281663e-07, + "loss": 0.9471, + "step": 6174 + }, + { + "epoch": 0.7424998496963867, + "grad_norm": 2.06949746265242, + "learning_rate": 6.559985224767801e-07, + "loss": 1.0206, + "step": 6175 + }, + { + "epoch": 0.7426200925870258, + "grad_norm": 2.503508109160147, + "learning_rate": 6.55421756609349e-07, + "loss": 0.9909, + "step": 6176 + }, + { + "epoch": 0.7427403354776649, + "grad_norm": 2.34172988299455, + "learning_rate": 6.54845194713369e-07, + "loss": 1.0248, + "step": 6177 + }, + { + "epoch": 0.742860578368304, + "grad_norm": 1.9995338969694925, + "learning_rate": 6.542688368763034e-07, + "loss": 1.0382, + "step": 6178 + }, + { + "epoch": 0.742980821258943, + "grad_norm": 1.6978475014420689, + "learning_rate": 6.536926831855854e-07, + "loss": 1.0053, + "step": 6179 + }, + { + "epoch": 0.7431010641495821, + "grad_norm": 2.9475566702612954, + "learning_rate": 6.531167337286165e-07, + "loss": 0.9697, + "step": 6180 + }, + { + "epoch": 0.7432213070402213, + "grad_norm": 1.5986281146130192, + "learning_rate": 6.52540988592768e-07, + "loss": 1.0366, + "step": 6181 + }, + { + "epoch": 0.7433415499308603, + "grad_norm": 2.104478705909544, + "learning_rate": 6.519654478653814e-07, + "loss": 1.0724, + "step": 6182 + }, + { + "epoch": 0.7434617928214994, + "grad_norm": 0.7633459073562237, + "learning_rate": 6.51390111633763e-07, + "loss": 0.7976, + "step": 6183 + }, + { + "epoch": 0.7435820357121385, + "grad_norm": 2.404185115097727, + "learning_rate": 6.508149799851932e-07, + "loss": 1.0004, + "step": 6184 + }, + { + "epoch": 0.7437022786027776, + "grad_norm": 2.07329485588718, + "learning_rate": 6.502400530069183e-07, + "loss": 0.8427, + "step": 6185 + }, + { + "epoch": 0.7438225214934167, + "grad_norm": 2.2030566049334763, + "learning_rate": 6.496653307861535e-07, + "loss": 0.9206, + "step": 6186 + }, + { + "epoch": 0.7439427643840558, + "grad_norm": 1.898120115774413, + "learning_rate": 6.490908134100857e-07, + "loss": 0.8955, + "step": 6187 + }, + { + "epoch": 0.7440630072746949, + "grad_norm": 2.0756502258734386, + "learning_rate": 6.48516500965866e-07, + "loss": 0.9308, + "step": 6188 + }, + { + "epoch": 0.7441832501653339, + "grad_norm": 1.8341541475766845, + "learning_rate": 6.479423935406192e-07, + "loss": 1.0548, + "step": 6189 + }, + { + "epoch": 0.7443034930559731, + "grad_norm": 0.8244666138153458, + "learning_rate": 6.473684912214357e-07, + "loss": 0.8581, + "step": 6190 + }, + { + "epoch": 0.7444237359466122, + "grad_norm": 2.3036911702005356, + "learning_rate": 6.467947940953778e-07, + "loss": 0.9324, + "step": 6191 + }, + { + "epoch": 0.7445439788372512, + "grad_norm": 1.768717934685123, + "learning_rate": 6.462213022494732e-07, + "loss": 0.9642, + "step": 6192 + }, + { + "epoch": 0.7446642217278904, + "grad_norm": 0.7801009692962223, + "learning_rate": 6.456480157707201e-07, + "loss": 0.8472, + "step": 6193 + }, + { + "epoch": 0.7447844646185294, + "grad_norm": 2.5848764507349213, + "learning_rate": 6.450749347460866e-07, + "loss": 1.0882, + "step": 6194 + }, + { + "epoch": 0.7449047075091685, + "grad_norm": 1.8788112053257422, + "learning_rate": 6.445020592625083e-07, + "loss": 1.0266, + "step": 6195 + }, + { + "epoch": 0.7450249503998077, + "grad_norm": 2.004694198157165, + "learning_rate": 6.4392938940689e-07, + "loss": 1.0422, + "step": 6196 + }, + { + "epoch": 0.7451451932904467, + "grad_norm": 3.036770385271789, + "learning_rate": 6.433569252661049e-07, + "loss": 0.9459, + "step": 6197 + }, + { + "epoch": 0.7452654361810858, + "grad_norm": 3.2329747214162388, + "learning_rate": 6.427846669269952e-07, + "loss": 0.9494, + "step": 6198 + }, + { + "epoch": 0.7453856790717249, + "grad_norm": 1.9913829899383912, + "learning_rate": 6.422126144763729e-07, + "loss": 1.0627, + "step": 6199 + }, + { + "epoch": 0.745505921962364, + "grad_norm": 2.1506114711186286, + "learning_rate": 6.416407680010174e-07, + "loss": 1.0037, + "step": 6200 + }, + { + "epoch": 0.745626164853003, + "grad_norm": 2.1431166355507156, + "learning_rate": 6.410691275876774e-07, + "loss": 1.0504, + "step": 6201 + }, + { + "epoch": 0.7457464077436422, + "grad_norm": 2.284300136774536, + "learning_rate": 6.404976933230704e-07, + "loss": 1.0029, + "step": 6202 + }, + { + "epoch": 0.7458666506342813, + "grad_norm": 2.6885815595415976, + "learning_rate": 6.399264652938813e-07, + "loss": 0.9611, + "step": 6203 + }, + { + "epoch": 0.7459868935249203, + "grad_norm": 2.380321563987908, + "learning_rate": 6.393554435867679e-07, + "loss": 0.9815, + "step": 6204 + }, + { + "epoch": 0.7461071364155595, + "grad_norm": 2.6403905098536367, + "learning_rate": 6.387846282883502e-07, + "loss": 1.0766, + "step": 6205 + }, + { + "epoch": 0.7462273793061985, + "grad_norm": 2.0258147072520964, + "learning_rate": 6.38214019485223e-07, + "loss": 1.0034, + "step": 6206 + }, + { + "epoch": 0.7463476221968376, + "grad_norm": 1.745084944791678, + "learning_rate": 6.376436172639461e-07, + "loss": 0.9533, + "step": 6207 + }, + { + "epoch": 0.7464678650874768, + "grad_norm": 2.727508643090306, + "learning_rate": 6.370734217110487e-07, + "loss": 0.8851, + "step": 6208 + }, + { + "epoch": 0.7465881079781158, + "grad_norm": 1.5228109000338692, + "learning_rate": 6.36503432913031e-07, + "loss": 0.8782, + "step": 6209 + }, + { + "epoch": 0.7467083508687549, + "grad_norm": 2.1353235239944377, + "learning_rate": 6.359336509563569e-07, + "loss": 0.9277, + "step": 6210 + }, + { + "epoch": 0.7468285937593939, + "grad_norm": 2.203846373719988, + "learning_rate": 6.353640759274641e-07, + "loss": 1.0454, + "step": 6211 + }, + { + "epoch": 0.7469488366500331, + "grad_norm": 2.8982706767397373, + "learning_rate": 6.347947079127556e-07, + "loss": 0.9828, + "step": 6212 + }, + { + "epoch": 0.7470690795406721, + "grad_norm": 2.618284809610727, + "learning_rate": 6.342255469986053e-07, + "loss": 1.0074, + "step": 6213 + }, + { + "epoch": 0.7471893224313112, + "grad_norm": 1.7829645839218262, + "learning_rate": 6.336565932713533e-07, + "loss": 1.0006, + "step": 6214 + }, + { + "epoch": 0.7473095653219504, + "grad_norm": 1.8423826675633357, + "learning_rate": 6.330878468173088e-07, + "loss": 1.0121, + "step": 6215 + }, + { + "epoch": 0.7474298082125894, + "grad_norm": 1.6856349952975662, + "learning_rate": 6.32519307722752e-07, + "loss": 0.9668, + "step": 6216 + }, + { + "epoch": 0.7475500511032285, + "grad_norm": 0.9419961561055734, + "learning_rate": 6.31950976073929e-07, + "loss": 0.7871, + "step": 6217 + }, + { + "epoch": 0.7476702939938676, + "grad_norm": 2.3459041614127574, + "learning_rate": 6.31382851957055e-07, + "loss": 1.0477, + "step": 6218 + }, + { + "epoch": 0.7477905368845067, + "grad_norm": 2.1035946235238403, + "learning_rate": 6.308149354583143e-07, + "loss": 0.9503, + "step": 6219 + }, + { + "epoch": 0.7479107797751458, + "grad_norm": 2.204659326753182, + "learning_rate": 6.302472266638586e-07, + "loss": 1.0546, + "step": 6220 + }, + { + "epoch": 0.7480310226657849, + "grad_norm": 2.1393045404116675, + "learning_rate": 6.296797256598101e-07, + "loss": 0.9329, + "step": 6221 + }, + { + "epoch": 0.748151265556424, + "grad_norm": 1.7941934230219125, + "learning_rate": 6.291124325322576e-07, + "loss": 1.0499, + "step": 6222 + }, + { + "epoch": 0.748271508447063, + "grad_norm": 1.682028483146238, + "learning_rate": 6.285453473672595e-07, + "loss": 0.8607, + "step": 6223 + }, + { + "epoch": 0.7483917513377022, + "grad_norm": 2.005637805739735, + "learning_rate": 6.279784702508415e-07, + "loss": 0.9884, + "step": 6224 + }, + { + "epoch": 0.7485119942283412, + "grad_norm": 0.7902020450150413, + "learning_rate": 6.274118012689979e-07, + "loss": 0.8235, + "step": 6225 + }, + { + "epoch": 0.7486322371189803, + "grad_norm": 1.4403203580066326, + "learning_rate": 6.268453405076943e-07, + "loss": 0.9178, + "step": 6226 + }, + { + "epoch": 0.7487524800096195, + "grad_norm": 2.015032185754174, + "learning_rate": 6.262790880528592e-07, + "loss": 1.0583, + "step": 6227 + }, + { + "epoch": 0.7488727229002585, + "grad_norm": 2.489704581358058, + "learning_rate": 6.257130439903951e-07, + "loss": 1.0325, + "step": 6228 + }, + { + "epoch": 0.7489929657908976, + "grad_norm": 1.9368147099563047, + "learning_rate": 6.251472084061695e-07, + "loss": 1.0507, + "step": 6229 + }, + { + "epoch": 0.7491132086815367, + "grad_norm": 2.099604151519665, + "learning_rate": 6.245815813860191e-07, + "loss": 1.1259, + "step": 6230 + }, + { + "epoch": 0.7492334515721758, + "grad_norm": 2.133761799024912, + "learning_rate": 6.240161630157495e-07, + "loss": 0.9384, + "step": 6231 + }, + { + "epoch": 0.7493536944628149, + "grad_norm": 2.144608960020485, + "learning_rate": 6.23450953381133e-07, + "loss": 0.935, + "step": 6232 + }, + { + "epoch": 0.749473937353454, + "grad_norm": 2.3798480495063155, + "learning_rate": 6.228859525679131e-07, + "loss": 0.9177, + "step": 6233 + }, + { + "epoch": 0.7495941802440931, + "grad_norm": 2.3787842767824765, + "learning_rate": 6.223211606617986e-07, + "loss": 1.0373, + "step": 6234 + }, + { + "epoch": 0.7497144231347321, + "grad_norm": 2.3712637775927807, + "learning_rate": 6.217565777484701e-07, + "loss": 1.0791, + "step": 6235 + }, + { + "epoch": 0.7498346660253713, + "grad_norm": 1.8238700190291732, + "learning_rate": 6.211922039135722e-07, + "loss": 1.04, + "step": 6236 + }, + { + "epoch": 0.7499549089160104, + "grad_norm": 1.8375006980635167, + "learning_rate": 6.206280392427201e-07, + "loss": 1.0482, + "step": 6237 + }, + { + "epoch": 0.7500751518066494, + "grad_norm": 1.7424198560194608, + "learning_rate": 6.200640838214983e-07, + "loss": 0.9754, + "step": 6238 + }, + { + "epoch": 0.7501953946972886, + "grad_norm": 1.9137850190489925, + "learning_rate": 6.195003377354578e-07, + "loss": 0.9056, + "step": 6239 + }, + { + "epoch": 0.7503156375879276, + "grad_norm": 2.4156838314891833, + "learning_rate": 6.189368010701183e-07, + "loss": 0.9674, + "step": 6240 + }, + { + "epoch": 0.7504358804785667, + "grad_norm": 2.2465813414372175, + "learning_rate": 6.183734739109683e-07, + "loss": 1.0018, + "step": 6241 + }, + { + "epoch": 0.7505561233692057, + "grad_norm": 2.248599146161269, + "learning_rate": 6.178103563434629e-07, + "loss": 0.9267, + "step": 6242 + }, + { + "epoch": 0.7506763662598449, + "grad_norm": 2.5789731566802314, + "learning_rate": 6.172474484530283e-07, + "loss": 1.0778, + "step": 6243 + }, + { + "epoch": 0.750796609150484, + "grad_norm": 1.7353107574250801, + "learning_rate": 6.166847503250563e-07, + "loss": 0.9951, + "step": 6244 + }, + { + "epoch": 0.750916852041123, + "grad_norm": 2.8745158243903286, + "learning_rate": 6.161222620449078e-07, + "loss": 1.0277, + "step": 6245 + }, + { + "epoch": 0.7510370949317622, + "grad_norm": 2.129002112233918, + "learning_rate": 6.155599836979117e-07, + "loss": 1.0394, + "step": 6246 + }, + { + "epoch": 0.7511573378224012, + "grad_norm": 2.4262425572213853, + "learning_rate": 6.149979153693649e-07, + "loss": 1.0547, + "step": 6247 + }, + { + "epoch": 0.7512775807130403, + "grad_norm": 1.9863149816927808, + "learning_rate": 6.144360571445343e-07, + "loss": 1.0054, + "step": 6248 + }, + { + "epoch": 0.7513978236036795, + "grad_norm": 2.176871938952458, + "learning_rate": 6.138744091086509e-07, + "loss": 1.0406, + "step": 6249 + }, + { + "epoch": 0.7515180664943185, + "grad_norm": 2.168073112106623, + "learning_rate": 6.133129713469183e-07, + "loss": 0.9669, + "step": 6250 + }, + { + "epoch": 0.7516383093849576, + "grad_norm": 5.849896304523837, + "learning_rate": 6.127517439445053e-07, + "loss": 0.873, + "step": 6251 + }, + { + "epoch": 0.7517585522755967, + "grad_norm": 2.026736365090216, + "learning_rate": 6.121907269865498e-07, + "loss": 1.0556, + "step": 6252 + }, + { + "epoch": 0.7518787951662358, + "grad_norm": 1.9955049306970603, + "learning_rate": 6.116299205581577e-07, + "loss": 0.91, + "step": 6253 + }, + { + "epoch": 0.7519990380568748, + "grad_norm": 2.028626376161973, + "learning_rate": 6.110693247444018e-07, + "loss": 0.9197, + "step": 6254 + }, + { + "epoch": 0.752119280947514, + "grad_norm": 2.5959880319445836, + "learning_rate": 6.105089396303258e-07, + "loss": 1.0603, + "step": 6255 + }, + { + "epoch": 0.7522395238381531, + "grad_norm": 1.786034499477024, + "learning_rate": 6.099487653009383e-07, + "loss": 0.9957, + "step": 6256 + }, + { + "epoch": 0.7523597667287921, + "grad_norm": 2.474088184897452, + "learning_rate": 6.093888018412192e-07, + "loss": 1.0721, + "step": 6257 + }, + { + "epoch": 0.7524800096194313, + "grad_norm": 0.7127779100747835, + "learning_rate": 6.088290493361125e-07, + "loss": 0.7842, + "step": 6258 + }, + { + "epoch": 0.7526002525100703, + "grad_norm": 2.06073338536825, + "learning_rate": 6.082695078705322e-07, + "loss": 0.953, + "step": 6259 + }, + { + "epoch": 0.7527204954007094, + "grad_norm": 2.2507621745791164, + "learning_rate": 6.077101775293618e-07, + "loss": 0.9232, + "step": 6260 + }, + { + "epoch": 0.7528407382913486, + "grad_norm": 2.422903322514934, + "learning_rate": 6.071510583974504e-07, + "loss": 1.0613, + "step": 6261 + }, + { + "epoch": 0.7529609811819876, + "grad_norm": 1.9450210291649916, + "learning_rate": 6.065921505596161e-07, + "loss": 0.9576, + "step": 6262 + }, + { + "epoch": 0.7530812240726267, + "grad_norm": 1.8883510818163611, + "learning_rate": 6.060334541006445e-07, + "loss": 1.0072, + "step": 6263 + }, + { + "epoch": 0.7532014669632658, + "grad_norm": 1.5729727156128614, + "learning_rate": 6.05474969105289e-07, + "loss": 0.9313, + "step": 6264 + }, + { + "epoch": 0.7533217098539049, + "grad_norm": 2.29152643179252, + "learning_rate": 6.049166956582725e-07, + "loss": 0.9709, + "step": 6265 + }, + { + "epoch": 0.753441952744544, + "grad_norm": 1.9963251565772249, + "learning_rate": 6.043586338442841e-07, + "loss": 1.1112, + "step": 6266 + }, + { + "epoch": 0.7535621956351831, + "grad_norm": 2.704404054242771, + "learning_rate": 6.038007837479815e-07, + "loss": 0.9696, + "step": 6267 + }, + { + "epoch": 0.7536824385258222, + "grad_norm": 2.114682563244951, + "learning_rate": 6.032431454539897e-07, + "loss": 0.8734, + "step": 6268 + }, + { + "epoch": 0.7538026814164612, + "grad_norm": 3.858548836563496, + "learning_rate": 6.026857190469014e-07, + "loss": 1.0519, + "step": 6269 + }, + { + "epoch": 0.7539229243071004, + "grad_norm": 3.1685332937082107, + "learning_rate": 6.0212850461128e-07, + "loss": 0.976, + "step": 6270 + }, + { + "epoch": 0.7540431671977395, + "grad_norm": 2.2050555272798578, + "learning_rate": 6.015715022316516e-07, + "loss": 0.9801, + "step": 6271 + }, + { + "epoch": 0.7541634100883785, + "grad_norm": 2.6557645466321818, + "learning_rate": 6.010147119925154e-07, + "loss": 1.0174, + "step": 6272 + }, + { + "epoch": 0.7542836529790176, + "grad_norm": 2.2143087935233763, + "learning_rate": 6.004581339783348e-07, + "loss": 0.9015, + "step": 6273 + }, + { + "epoch": 0.7544038958696567, + "grad_norm": 4.1978626989235455, + "learning_rate": 5.999017682735425e-07, + "loss": 0.9198, + "step": 6274 + }, + { + "epoch": 0.7545241387602958, + "grad_norm": 2.2118049779944777, + "learning_rate": 5.993456149625387e-07, + "loss": 0.8991, + "step": 6275 + }, + { + "epoch": 0.7546443816509348, + "grad_norm": 2.0379194658668753, + "learning_rate": 5.987896741296909e-07, + "loss": 1.0582, + "step": 6276 + }, + { + "epoch": 0.754764624541574, + "grad_norm": 2.103521745508157, + "learning_rate": 5.982339458593361e-07, + "loss": 1.0205, + "step": 6277 + }, + { + "epoch": 0.7548848674322131, + "grad_norm": 1.5811714002049784, + "learning_rate": 5.976784302357767e-07, + "loss": 1.0768, + "step": 6278 + }, + { + "epoch": 0.7550051103228521, + "grad_norm": 2.255175360481898, + "learning_rate": 5.971231273432855e-07, + "loss": 0.9675, + "step": 6279 + }, + { + "epoch": 0.7551253532134913, + "grad_norm": 0.8171753249711093, + "learning_rate": 5.965680372661e-07, + "loss": 0.7827, + "step": 6280 + }, + { + "epoch": 0.7552455961041303, + "grad_norm": 1.8239436136113267, + "learning_rate": 5.960131600884266e-07, + "loss": 0.8015, + "step": 6281 + }, + { + "epoch": 0.7553658389947694, + "grad_norm": 1.7415730714365896, + "learning_rate": 5.954584958944413e-07, + "loss": 0.9957, + "step": 6282 + }, + { + "epoch": 0.7554860818854086, + "grad_norm": 3.256204112016968, + "learning_rate": 5.949040447682854e-07, + "loss": 1.0557, + "step": 6283 + }, + { + "epoch": 0.7556063247760476, + "grad_norm": 2.0173644610968253, + "learning_rate": 5.943498067940686e-07, + "loss": 0.92, + "step": 6284 + }, + { + "epoch": 0.7557265676666867, + "grad_norm": 2.743925842350757, + "learning_rate": 5.937957820558686e-07, + "loss": 1.0545, + "step": 6285 + }, + { + "epoch": 0.7558468105573258, + "grad_norm": 0.8550327182666685, + "learning_rate": 5.932419706377296e-07, + "loss": 0.8911, + "step": 6286 + }, + { + "epoch": 0.7559670534479649, + "grad_norm": 1.9486924189000154, + "learning_rate": 5.92688372623666e-07, + "loss": 0.9797, + "step": 6287 + }, + { + "epoch": 0.7560872963386039, + "grad_norm": 2.559433195130217, + "learning_rate": 5.921349880976574e-07, + "loss": 0.9744, + "step": 6288 + }, + { + "epoch": 0.7562075392292431, + "grad_norm": 1.6998727526444752, + "learning_rate": 5.915818171436515e-07, + "loss": 1.0541, + "step": 6289 + }, + { + "epoch": 0.7563277821198822, + "grad_norm": 1.8479781727265556, + "learning_rate": 5.910288598455642e-07, + "loss": 0.9806, + "step": 6290 + }, + { + "epoch": 0.7564480250105212, + "grad_norm": 2.698190937805, + "learning_rate": 5.90476116287278e-07, + "loss": 0.9805, + "step": 6291 + }, + { + "epoch": 0.7565682679011604, + "grad_norm": 1.9016039778762892, + "learning_rate": 5.899235865526456e-07, + "loss": 0.9179, + "step": 6292 + }, + { + "epoch": 0.7566885107917994, + "grad_norm": 1.7016639391995254, + "learning_rate": 5.893712707254825e-07, + "loss": 1.0595, + "step": 6293 + }, + { + "epoch": 0.7568087536824385, + "grad_norm": 2.581922474011216, + "learning_rate": 5.888191688895769e-07, + "loss": 0.8959, + "step": 6294 + }, + { + "epoch": 0.7569289965730777, + "grad_norm": 2.0904353079960414, + "learning_rate": 5.882672811286813e-07, + "loss": 0.8587, + "step": 6295 + }, + { + "epoch": 0.7570492394637167, + "grad_norm": 1.970354183521457, + "learning_rate": 5.877156075265166e-07, + "loss": 0.9267, + "step": 6296 + }, + { + "epoch": 0.7571694823543558, + "grad_norm": 2.6171729243997115, + "learning_rate": 5.871641481667715e-07, + "loss": 0.9333, + "step": 6297 + }, + { + "epoch": 0.7572897252449949, + "grad_norm": 1.6952047006112527, + "learning_rate": 5.866129031331011e-07, + "loss": 1.0804, + "step": 6298 + }, + { + "epoch": 0.757409968135634, + "grad_norm": 2.1580972913106136, + "learning_rate": 5.8606187250913e-07, + "loss": 1.0725, + "step": 6299 + }, + { + "epoch": 0.757530211026273, + "grad_norm": 2.222110299099927, + "learning_rate": 5.855110563784482e-07, + "loss": 1.0801, + "step": 6300 + }, + { + "epoch": 0.7576504539169122, + "grad_norm": 1.655111533379925, + "learning_rate": 5.849604548246156e-07, + "loss": 0.8793, + "step": 6301 + }, + { + "epoch": 0.7577706968075513, + "grad_norm": 2.07012443408045, + "learning_rate": 5.844100679311565e-07, + "loss": 1.0405, + "step": 6302 + }, + { + "epoch": 0.7578909396981903, + "grad_norm": 2.065757029320489, + "learning_rate": 5.838598957815637e-07, + "loss": 0.9992, + "step": 6303 + }, + { + "epoch": 0.7580111825888295, + "grad_norm": 1.5175110398319884, + "learning_rate": 5.833099384592996e-07, + "loss": 1.0891, + "step": 6304 + }, + { + "epoch": 0.7581314254794685, + "grad_norm": 2.034274153546446, + "learning_rate": 5.827601960477913e-07, + "loss": 0.9496, + "step": 6305 + }, + { + "epoch": 0.7582516683701076, + "grad_norm": 1.826610100536265, + "learning_rate": 5.822106686304344e-07, + "loss": 0.944, + "step": 6306 + }, + { + "epoch": 0.7583719112607467, + "grad_norm": 1.8512102910740758, + "learning_rate": 5.816613562905919e-07, + "loss": 0.8124, + "step": 6307 + }, + { + "epoch": 0.7584921541513858, + "grad_norm": 1.5731198093357468, + "learning_rate": 5.811122591115933e-07, + "loss": 0.9364, + "step": 6308 + }, + { + "epoch": 0.7586123970420249, + "grad_norm": 2.3028525078702176, + "learning_rate": 5.805633771767376e-07, + "loss": 0.9541, + "step": 6309 + }, + { + "epoch": 0.7587326399326639, + "grad_norm": 1.9859801881345553, + "learning_rate": 5.800147105692888e-07, + "loss": 1.0152, + "step": 6310 + }, + { + "epoch": 0.7588528828233031, + "grad_norm": 1.8782383610032083, + "learning_rate": 5.794662593724795e-07, + "loss": 1.0283, + "step": 6311 + }, + { + "epoch": 0.7589731257139422, + "grad_norm": 2.080371646140582, + "learning_rate": 5.789180236695091e-07, + "loss": 0.9875, + "step": 6312 + }, + { + "epoch": 0.7590933686045812, + "grad_norm": 1.8215096265992397, + "learning_rate": 5.78370003543544e-07, + "loss": 1.0868, + "step": 6313 + }, + { + "epoch": 0.7592136114952204, + "grad_norm": 2.514386458592449, + "learning_rate": 5.778221990777203e-07, + "loss": 1.0746, + "step": 6314 + }, + { + "epoch": 0.7593338543858594, + "grad_norm": 3.5505726903047887, + "learning_rate": 5.772746103551372e-07, + "loss": 1.0685, + "step": 6315 + }, + { + "epoch": 0.7594540972764985, + "grad_norm": 2.67248438974216, + "learning_rate": 5.767272374588648e-07, + "loss": 0.953, + "step": 6316 + }, + { + "epoch": 0.7595743401671377, + "grad_norm": 3.459662553505885, + "learning_rate": 5.76180080471939e-07, + "loss": 1.019, + "step": 6317 + }, + { + "epoch": 0.7596945830577767, + "grad_norm": 2.576937315208438, + "learning_rate": 5.756331394773631e-07, + "loss": 0.9599, + "step": 6318 + }, + { + "epoch": 0.7598148259484158, + "grad_norm": 1.9266951986309964, + "learning_rate": 5.750864145581071e-07, + "loss": 0.9978, + "step": 6319 + }, + { + "epoch": 0.7599350688390549, + "grad_norm": 2.39483929343788, + "learning_rate": 5.745399057971085e-07, + "loss": 1.0957, + "step": 6320 + }, + { + "epoch": 0.760055311729694, + "grad_norm": 1.99675182983498, + "learning_rate": 5.739936132772738e-07, + "loss": 0.988, + "step": 6321 + }, + { + "epoch": 0.760175554620333, + "grad_norm": 2.20869877028841, + "learning_rate": 5.734475370814733e-07, + "loss": 0.9806, + "step": 6322 + }, + { + "epoch": 0.7602957975109722, + "grad_norm": 1.6642723846603409, + "learning_rate": 5.729016772925483e-07, + "loss": 1.0232, + "step": 6323 + }, + { + "epoch": 0.7604160404016113, + "grad_norm": 2.124603122225757, + "learning_rate": 5.723560339933038e-07, + "loss": 0.9419, + "step": 6324 + }, + { + "epoch": 0.7605362832922503, + "grad_norm": 1.9590549627424694, + "learning_rate": 5.71810607266513e-07, + "loss": 0.8888, + "step": 6325 + }, + { + "epoch": 0.7606565261828895, + "grad_norm": 1.8479921708993912, + "learning_rate": 5.712653971949184e-07, + "loss": 0.8448, + "step": 6326 + }, + { + "epoch": 0.7607767690735285, + "grad_norm": 2.648880732983885, + "learning_rate": 5.707204038612268e-07, + "loss": 0.992, + "step": 6327 + }, + { + "epoch": 0.7608970119641676, + "grad_norm": 2.0388728378777436, + "learning_rate": 5.701756273481138e-07, + "loss": 0.9734, + "step": 6328 + }, + { + "epoch": 0.7610172548548068, + "grad_norm": 1.627640998932206, + "learning_rate": 5.696310677382212e-07, + "loss": 0.977, + "step": 6329 + }, + { + "epoch": 0.7611374977454458, + "grad_norm": 0.8709397457820652, + "learning_rate": 5.690867251141576e-07, + "loss": 0.8551, + "step": 6330 + }, + { + "epoch": 0.7612577406360849, + "grad_norm": 2.35727481968066, + "learning_rate": 5.685425995585013e-07, + "loss": 1.1512, + "step": 6331 + }, + { + "epoch": 0.761377983526724, + "grad_norm": 0.77106447019323, + "learning_rate": 5.679986911537935e-07, + "loss": 0.827, + "step": 6332 + }, + { + "epoch": 0.7614982264173631, + "grad_norm": 2.052777000442011, + "learning_rate": 5.674549999825462e-07, + "loss": 0.9091, + "step": 6333 + }, + { + "epoch": 0.7616184693080021, + "grad_norm": 0.9221962191338554, + "learning_rate": 5.669115261272363e-07, + "loss": 0.9516, + "step": 6334 + }, + { + "epoch": 0.7617387121986413, + "grad_norm": 2.4518767661759986, + "learning_rate": 5.663682696703081e-07, + "loss": 0.9628, + "step": 6335 + }, + { + "epoch": 0.7618589550892804, + "grad_norm": 1.9460860097173012, + "learning_rate": 5.658252306941746e-07, + "loss": 1.0589, + "step": 6336 + }, + { + "epoch": 0.7619791979799194, + "grad_norm": 2.0088840104262347, + "learning_rate": 5.65282409281212e-07, + "loss": 0.9883, + "step": 6337 + }, + { + "epoch": 0.7620994408705585, + "grad_norm": 2.099651389477888, + "learning_rate": 5.64739805513768e-07, + "loss": 0.9355, + "step": 6338 + }, + { + "epoch": 0.7622196837611976, + "grad_norm": 0.7992948926525051, + "learning_rate": 5.641974194741541e-07, + "loss": 0.7946, + "step": 6339 + }, + { + "epoch": 0.7623399266518367, + "grad_norm": 0.7979941269914204, + "learning_rate": 5.636552512446502e-07, + "loss": 0.8747, + "step": 6340 + }, + { + "epoch": 0.7624601695424758, + "grad_norm": 1.7378202019808728, + "learning_rate": 5.631133009075027e-07, + "loss": 1.0165, + "step": 6341 + }, + { + "epoch": 0.7625804124331149, + "grad_norm": 2.1740958296598127, + "learning_rate": 5.625715685449242e-07, + "loss": 0.9252, + "step": 6342 + }, + { + "epoch": 0.762700655323754, + "grad_norm": 1.952783417395884, + "learning_rate": 5.620300542390966e-07, + "loss": 0.9519, + "step": 6343 + }, + { + "epoch": 0.762820898214393, + "grad_norm": 1.9305146332694756, + "learning_rate": 5.614887580721659e-07, + "loss": 1.0884, + "step": 6344 + }, + { + "epoch": 0.7629411411050322, + "grad_norm": 2.0786501572258986, + "learning_rate": 5.609476801262481e-07, + "loss": 0.9742, + "step": 6345 + }, + { + "epoch": 0.7630613839956712, + "grad_norm": 2.5709965963678667, + "learning_rate": 5.604068204834223e-07, + "loss": 0.8787, + "step": 6346 + }, + { + "epoch": 0.7631816268863103, + "grad_norm": 2.219550512190649, + "learning_rate": 5.598661792257367e-07, + "loss": 1.004, + "step": 6347 + }, + { + "epoch": 0.7633018697769495, + "grad_norm": 2.722279679952776, + "learning_rate": 5.593257564352071e-07, + "loss": 0.9975, + "step": 6348 + }, + { + "epoch": 0.7634221126675885, + "grad_norm": 1.6074345566167048, + "learning_rate": 5.58785552193815e-07, + "loss": 0.9947, + "step": 6349 + }, + { + "epoch": 0.7635423555582276, + "grad_norm": 1.9258191954180308, + "learning_rate": 5.582455665835086e-07, + "loss": 0.9944, + "step": 6350 + }, + { + "epoch": 0.7636625984488667, + "grad_norm": 3.3209311234368184, + "learning_rate": 5.577057996862036e-07, + "loss": 0.9657, + "step": 6351 + }, + { + "epoch": 0.7637828413395058, + "grad_norm": 1.8035215103816822, + "learning_rate": 5.571662515837814e-07, + "loss": 0.9961, + "step": 6352 + }, + { + "epoch": 0.7639030842301449, + "grad_norm": 2.3171550190270285, + "learning_rate": 5.566269223580926e-07, + "loss": 1.0753, + "step": 6353 + }, + { + "epoch": 0.764023327120784, + "grad_norm": 1.8590338257723018, + "learning_rate": 5.560878120909511e-07, + "loss": 0.9896, + "step": 6354 + }, + { + "epoch": 0.7641435700114231, + "grad_norm": 0.8581800302307228, + "learning_rate": 5.55548920864141e-07, + "loss": 0.8231, + "step": 6355 + }, + { + "epoch": 0.7642638129020621, + "grad_norm": 1.960010263746986, + "learning_rate": 5.550102487594113e-07, + "loss": 1.0188, + "step": 6356 + }, + { + "epoch": 0.7643840557927013, + "grad_norm": 1.7184595035934722, + "learning_rate": 5.54471795858477e-07, + "loss": 0.9494, + "step": 6357 + }, + { + "epoch": 0.7645042986833404, + "grad_norm": 2.0984710304382044, + "learning_rate": 5.539335622430235e-07, + "loss": 1.0614, + "step": 6358 + }, + { + "epoch": 0.7646245415739794, + "grad_norm": 2.051664151464677, + "learning_rate": 5.533955479946975e-07, + "loss": 0.9828, + "step": 6359 + }, + { + "epoch": 0.7647447844646186, + "grad_norm": 0.8679205742899271, + "learning_rate": 5.528577531951173e-07, + "loss": 0.8949, + "step": 6360 + }, + { + "epoch": 0.7648650273552576, + "grad_norm": 2.7490095609108027, + "learning_rate": 5.523201779258653e-07, + "loss": 0.9772, + "step": 6361 + }, + { + "epoch": 0.7649852702458967, + "grad_norm": 2.078550711028983, + "learning_rate": 5.517828222684912e-07, + "loss": 1.0779, + "step": 6362 + }, + { + "epoch": 0.7651055131365359, + "grad_norm": 0.7786304808328967, + "learning_rate": 5.512456863045117e-07, + "loss": 0.8285, + "step": 6363 + }, + { + "epoch": 0.7652257560271749, + "grad_norm": 1.9692137641241283, + "learning_rate": 5.507087701154089e-07, + "loss": 0.9822, + "step": 6364 + }, + { + "epoch": 0.765345998917814, + "grad_norm": 1.8602072512104812, + "learning_rate": 5.50172073782634e-07, + "loss": 0.9932, + "step": 6365 + }, + { + "epoch": 0.7654662418084531, + "grad_norm": 1.8202276251001506, + "learning_rate": 5.496355973876023e-07, + "loss": 1.1142, + "step": 6366 + }, + { + "epoch": 0.7655864846990922, + "grad_norm": 2.1362912379749366, + "learning_rate": 5.490993410116984e-07, + "loss": 0.9446, + "step": 6367 + }, + { + "epoch": 0.7657067275897312, + "grad_norm": 1.8886531288455333, + "learning_rate": 5.485633047362704e-07, + "loss": 0.9319, + "step": 6368 + }, + { + "epoch": 0.7658269704803703, + "grad_norm": 2.3154792152877217, + "learning_rate": 5.480274886426341e-07, + "loss": 1.0285, + "step": 6369 + }, + { + "epoch": 0.7659472133710095, + "grad_norm": 2.1417369321386825, + "learning_rate": 5.474918928120744e-07, + "loss": 1.0167, + "step": 6370 + }, + { + "epoch": 0.7660674562616485, + "grad_norm": 2.0736654517084308, + "learning_rate": 5.469565173258392e-07, + "loss": 1.11, + "step": 6371 + }, + { + "epoch": 0.7661876991522876, + "grad_norm": 1.7031165306510676, + "learning_rate": 5.464213622651454e-07, + "loss": 0.8754, + "step": 6372 + }, + { + "epoch": 0.7663079420429267, + "grad_norm": 1.7106849100746189, + "learning_rate": 5.458864277111753e-07, + "loss": 1.0825, + "step": 6373 + }, + { + "epoch": 0.7664281849335658, + "grad_norm": 3.871477802830286, + "learning_rate": 5.453517137450769e-07, + "loss": 0.9305, + "step": 6374 + }, + { + "epoch": 0.7665484278242048, + "grad_norm": 1.8808748397112156, + "learning_rate": 5.448172204479684e-07, + "loss": 0.998, + "step": 6375 + }, + { + "epoch": 0.766668670714844, + "grad_norm": 1.857175086601118, + "learning_rate": 5.442829479009294e-07, + "loss": 0.9836, + "step": 6376 + }, + { + "epoch": 0.7667889136054831, + "grad_norm": 2.0514740273311904, + "learning_rate": 5.437488961850103e-07, + "loss": 0.9506, + "step": 6377 + }, + { + "epoch": 0.7669091564961221, + "grad_norm": 1.8889064071191979, + "learning_rate": 5.432150653812258e-07, + "loss": 0.9916, + "step": 6378 + }, + { + "epoch": 0.7670293993867613, + "grad_norm": 3.108952019066932, + "learning_rate": 5.42681455570557e-07, + "loss": 1.0659, + "step": 6379 + }, + { + "epoch": 0.7671496422774003, + "grad_norm": 2.211723053286943, + "learning_rate": 5.42148066833954e-07, + "loss": 0.8838, + "step": 6380 + }, + { + "epoch": 0.7672698851680394, + "grad_norm": 2.0973675349786043, + "learning_rate": 5.416148992523289e-07, + "loss": 0.9909, + "step": 6381 + }, + { + "epoch": 0.7673901280586786, + "grad_norm": 2.004028911442939, + "learning_rate": 5.410819529065644e-07, + "loss": 1.0243, + "step": 6382 + }, + { + "epoch": 0.7675103709493176, + "grad_norm": 1.9883618291628602, + "learning_rate": 5.405492278775079e-07, + "loss": 0.8903, + "step": 6383 + }, + { + "epoch": 0.7676306138399567, + "grad_norm": 2.4820337359758464, + "learning_rate": 5.400167242459732e-07, + "loss": 1.0314, + "step": 6384 + }, + { + "epoch": 0.7677508567305958, + "grad_norm": 1.8500194316565202, + "learning_rate": 5.394844420927405e-07, + "loss": 1.0428, + "step": 6385 + }, + { + "epoch": 0.7678710996212349, + "grad_norm": 3.1601038906064893, + "learning_rate": 5.389523814985562e-07, + "loss": 0.9702, + "step": 6386 + }, + { + "epoch": 0.767991342511874, + "grad_norm": 1.7831162831614524, + "learning_rate": 5.384205425441344e-07, + "loss": 0.9968, + "step": 6387 + }, + { + "epoch": 0.7681115854025131, + "grad_norm": 1.733860936318273, + "learning_rate": 5.378889253101537e-07, + "loss": 1.0789, + "step": 6388 + }, + { + "epoch": 0.7682318282931522, + "grad_norm": 1.7335614882600765, + "learning_rate": 5.373575298772617e-07, + "loss": 1.0457, + "step": 6389 + }, + { + "epoch": 0.7683520711837912, + "grad_norm": 0.7665056107845106, + "learning_rate": 5.368263563260689e-07, + "loss": 0.8513, + "step": 6390 + }, + { + "epoch": 0.7684723140744304, + "grad_norm": 1.6697793821698808, + "learning_rate": 5.362954047371537e-07, + "loss": 0.8767, + "step": 6391 + }, + { + "epoch": 0.7685925569650695, + "grad_norm": 2.143364594183606, + "learning_rate": 5.357646751910627e-07, + "loss": 0.9581, + "step": 6392 + }, + { + "epoch": 0.7687127998557085, + "grad_norm": 2.4050305732163717, + "learning_rate": 5.352341677683061e-07, + "loss": 1.0351, + "step": 6393 + }, + { + "epoch": 0.7688330427463477, + "grad_norm": 3.4569853181535497, + "learning_rate": 5.347038825493617e-07, + "loss": 1.0291, + "step": 6394 + }, + { + "epoch": 0.7689532856369867, + "grad_norm": 1.9011571422325062, + "learning_rate": 5.341738196146732e-07, + "loss": 0.9211, + "step": 6395 + }, + { + "epoch": 0.7690735285276258, + "grad_norm": 2.261201209643216, + "learning_rate": 5.336439790446503e-07, + "loss": 0.9732, + "step": 6396 + }, + { + "epoch": 0.769193771418265, + "grad_norm": 1.778317698142127, + "learning_rate": 5.331143609196711e-07, + "loss": 0.8624, + "step": 6397 + }, + { + "epoch": 0.769314014308904, + "grad_norm": 2.6191332516150405, + "learning_rate": 5.325849653200758e-07, + "loss": 1.0077, + "step": 6398 + }, + { + "epoch": 0.7694342571995431, + "grad_norm": 1.939623438184576, + "learning_rate": 5.32055792326175e-07, + "loss": 1.0029, + "step": 6399 + }, + { + "epoch": 0.7695545000901821, + "grad_norm": 2.7007526726130826, + "learning_rate": 5.315268420182437e-07, + "loss": 0.9677, + "step": 6400 + }, + { + "epoch": 0.7696747429808213, + "grad_norm": 2.116415072929999, + "learning_rate": 5.309981144765221e-07, + "loss": 0.9992, + "step": 6401 + }, + { + "epoch": 0.7697949858714603, + "grad_norm": 2.2627918287032918, + "learning_rate": 5.304696097812196e-07, + "loss": 0.9874, + "step": 6402 + }, + { + "epoch": 0.7699152287620994, + "grad_norm": 3.5020632792996813, + "learning_rate": 5.299413280125078e-07, + "loss": 0.8327, + "step": 6403 + }, + { + "epoch": 0.7700354716527386, + "grad_norm": 1.9621832184943062, + "learning_rate": 5.294132692505284e-07, + "loss": 0.9657, + "step": 6404 + }, + { + "epoch": 0.7701557145433776, + "grad_norm": 1.9309974579362221, + "learning_rate": 5.288854335753861e-07, + "loss": 1.0242, + "step": 6405 + }, + { + "epoch": 0.7702759574340167, + "grad_norm": 1.9351012855458096, + "learning_rate": 5.283578210671551e-07, + "loss": 0.993, + "step": 6406 + }, + { + "epoch": 0.7703962003246558, + "grad_norm": 2.562301441966654, + "learning_rate": 5.278304318058719e-07, + "loss": 1.0009, + "step": 6407 + }, + { + "epoch": 0.7705164432152949, + "grad_norm": 4.232698340863777, + "learning_rate": 5.273032658715411e-07, + "loss": 1.0256, + "step": 6408 + }, + { + "epoch": 0.7706366861059339, + "grad_norm": 1.7943494213889386, + "learning_rate": 5.267763233441347e-07, + "loss": 1.0004, + "step": 6409 + }, + { + "epoch": 0.7707569289965731, + "grad_norm": 2.2982749565111478, + "learning_rate": 5.26249604303588e-07, + "loss": 0.9311, + "step": 6410 + }, + { + "epoch": 0.7708771718872122, + "grad_norm": 2.0016247587028144, + "learning_rate": 5.257231088298057e-07, + "loss": 1.0198, + "step": 6411 + }, + { + "epoch": 0.7709974147778512, + "grad_norm": 0.8097709158739733, + "learning_rate": 5.25196837002655e-07, + "loss": 0.7764, + "step": 6412 + }, + { + "epoch": 0.7711176576684904, + "grad_norm": 2.2921613217180576, + "learning_rate": 5.24670788901971e-07, + "loss": 0.9184, + "step": 6413 + }, + { + "epoch": 0.7712379005591294, + "grad_norm": 2.2077883881739133, + "learning_rate": 5.241449646075557e-07, + "loss": 0.9214, + "step": 6414 + }, + { + "epoch": 0.7713581434497685, + "grad_norm": 2.3591380221362663, + "learning_rate": 5.236193641991762e-07, + "loss": 0.9633, + "step": 6415 + }, + { + "epoch": 0.7714783863404077, + "grad_norm": 6.067703373679125, + "learning_rate": 5.23093987756565e-07, + "loss": 0.937, + "step": 6416 + }, + { + "epoch": 0.7715986292310467, + "grad_norm": 2.48577706942014, + "learning_rate": 5.225688353594217e-07, + "loss": 0.9871, + "step": 6417 + }, + { + "epoch": 0.7717188721216858, + "grad_norm": 2.463871638732107, + "learning_rate": 5.220439070874108e-07, + "loss": 1.0103, + "step": 6418 + }, + { + "epoch": 0.7718391150123249, + "grad_norm": 1.6106864992656473, + "learning_rate": 5.215192030201652e-07, + "loss": 0.9479, + "step": 6419 + }, + { + "epoch": 0.771959357902964, + "grad_norm": 1.8504319099199846, + "learning_rate": 5.209947232372798e-07, + "loss": 1.1022, + "step": 6420 + }, + { + "epoch": 0.772079600793603, + "grad_norm": 1.8533009926400728, + "learning_rate": 5.204704678183196e-07, + "loss": 1.0464, + "step": 6421 + }, + { + "epoch": 0.7721998436842422, + "grad_norm": 1.987167556117915, + "learning_rate": 5.19946436842813e-07, + "loss": 1.0915, + "step": 6422 + }, + { + "epoch": 0.7723200865748813, + "grad_norm": 1.6480532374579056, + "learning_rate": 5.194226303902546e-07, + "loss": 0.9226, + "step": 6423 + }, + { + "epoch": 0.7724403294655203, + "grad_norm": 2.1138425635086406, + "learning_rate": 5.188990485401072e-07, + "loss": 0.9449, + "step": 6424 + }, + { + "epoch": 0.7725605723561595, + "grad_norm": 2.2322971312479294, + "learning_rate": 5.183756913717954e-07, + "loss": 1.101, + "step": 6425 + }, + { + "epoch": 0.7726808152467985, + "grad_norm": 3.407281981761249, + "learning_rate": 5.178525589647136e-07, + "loss": 0.9659, + "step": 6426 + }, + { + "epoch": 0.7728010581374376, + "grad_norm": 2.3644419444380875, + "learning_rate": 5.173296513982197e-07, + "loss": 1.0235, + "step": 6427 + }, + { + "epoch": 0.7729213010280768, + "grad_norm": 3.2793870269397813, + "learning_rate": 5.168069687516398e-07, + "loss": 0.8881, + "step": 6428 + }, + { + "epoch": 0.7730415439187158, + "grad_norm": 2.0172796037775464, + "learning_rate": 5.16284511104263e-07, + "loss": 0.9557, + "step": 6429 + }, + { + "epoch": 0.7731617868093549, + "grad_norm": 3.3834075756728033, + "learning_rate": 5.157622785353457e-07, + "loss": 1.0463, + "step": 6430 + }, + { + "epoch": 0.7732820296999939, + "grad_norm": 0.6532730746903876, + "learning_rate": 5.152402711241113e-07, + "loss": 0.8419, + "step": 6431 + }, + { + "epoch": 0.7734022725906331, + "grad_norm": 1.8787919166340699, + "learning_rate": 5.147184889497465e-07, + "loss": 1.0653, + "step": 6432 + }, + { + "epoch": 0.7735225154812722, + "grad_norm": 8.623136899657478, + "learning_rate": 5.141969320914072e-07, + "loss": 1.0363, + "step": 6433 + }, + { + "epoch": 0.7736427583719112, + "grad_norm": 4.886838184241918, + "learning_rate": 5.136756006282113e-07, + "loss": 0.8565, + "step": 6434 + }, + { + "epoch": 0.7737630012625504, + "grad_norm": 2.487697562801987, + "learning_rate": 5.131544946392446e-07, + "loss": 1.0887, + "step": 6435 + }, + { + "epoch": 0.7738832441531894, + "grad_norm": 2.057551371604012, + "learning_rate": 5.126336142035592e-07, + "loss": 0.8758, + "step": 6436 + }, + { + "epoch": 0.7740034870438285, + "grad_norm": 2.6877889256086065, + "learning_rate": 5.121129594001721e-07, + "loss": 0.9633, + "step": 6437 + }, + { + "epoch": 0.7741237299344677, + "grad_norm": 1.5916927748264784, + "learning_rate": 5.115925303080661e-07, + "loss": 1.0488, + "step": 6438 + }, + { + "epoch": 0.7742439728251067, + "grad_norm": 2.600085323107487, + "learning_rate": 5.110723270061899e-07, + "loss": 1.029, + "step": 6439 + }, + { + "epoch": 0.7743642157157458, + "grad_norm": 2.2624470488532427, + "learning_rate": 5.105523495734572e-07, + "loss": 1.0315, + "step": 6440 + }, + { + "epoch": 0.7744844586063849, + "grad_norm": 1.5901478469535146, + "learning_rate": 5.100325980887499e-07, + "loss": 0.9889, + "step": 6441 + }, + { + "epoch": 0.774604701497024, + "grad_norm": 1.89897066643484, + "learning_rate": 5.095130726309116e-07, + "loss": 1.0726, + "step": 6442 + }, + { + "epoch": 0.774724944387663, + "grad_norm": 0.7941406685322057, + "learning_rate": 5.089937732787559e-07, + "loss": 0.8274, + "step": 6443 + }, + { + "epoch": 0.7748451872783022, + "grad_norm": 2.7105026555965765, + "learning_rate": 5.084747001110592e-07, + "loss": 0.9034, + "step": 6444 + }, + { + "epoch": 0.7749654301689413, + "grad_norm": 1.7232674619431092, + "learning_rate": 5.07955853206564e-07, + "loss": 0.9396, + "step": 6445 + }, + { + "epoch": 0.7750856730595803, + "grad_norm": 3.679388791410461, + "learning_rate": 5.074372326439807e-07, + "loss": 0.9465, + "step": 6446 + }, + { + "epoch": 0.7752059159502195, + "grad_norm": 2.112016846069428, + "learning_rate": 5.069188385019814e-07, + "loss": 0.9673, + "step": 6447 + }, + { + "epoch": 0.7753261588408585, + "grad_norm": 3.7952907695124605, + "learning_rate": 5.064006708592077e-07, + "loss": 0.8422, + "step": 6448 + }, + { + "epoch": 0.7754464017314976, + "grad_norm": 2.4511813098984256, + "learning_rate": 5.058827297942641e-07, + "loss": 0.9977, + "step": 6449 + }, + { + "epoch": 0.7755666446221368, + "grad_norm": 1.8084328499525426, + "learning_rate": 5.053650153857237e-07, + "loss": 0.9863, + "step": 6450 + }, + { + "epoch": 0.7756868875127758, + "grad_norm": 2.324603701215822, + "learning_rate": 5.048475277121214e-07, + "loss": 0.9359, + "step": 6451 + }, + { + "epoch": 0.7758071304034149, + "grad_norm": 1.8517134077809299, + "learning_rate": 5.043302668519598e-07, + "loss": 1.0083, + "step": 6452 + }, + { + "epoch": 0.775927373294054, + "grad_norm": 1.8788376635050335, + "learning_rate": 5.038132328837079e-07, + "loss": 0.9556, + "step": 6453 + }, + { + "epoch": 0.7760476161846931, + "grad_norm": 2.0384375538958293, + "learning_rate": 5.032964258857993e-07, + "loss": 0.9725, + "step": 6454 + }, + { + "epoch": 0.7761678590753321, + "grad_norm": 1.5660975430292414, + "learning_rate": 5.027798459366329e-07, + "loss": 0.9197, + "step": 6455 + }, + { + "epoch": 0.7762881019659713, + "grad_norm": 1.5489483434290416, + "learning_rate": 5.02263493114573e-07, + "loss": 0.8732, + "step": 6456 + }, + { + "epoch": 0.7764083448566104, + "grad_norm": 3.2087986266044544, + "learning_rate": 5.017473674979502e-07, + "loss": 1.005, + "step": 6457 + }, + { + "epoch": 0.7765285877472494, + "grad_norm": 0.7504343920627699, + "learning_rate": 5.01231469165061e-07, + "loss": 0.8206, + "step": 6458 + }, + { + "epoch": 0.7766488306378886, + "grad_norm": 0.8619872531722846, + "learning_rate": 5.007157981941663e-07, + "loss": 0.8067, + "step": 6459 + }, + { + "epoch": 0.7767690735285276, + "grad_norm": 0.8868050890325283, + "learning_rate": 5.002003546634928e-07, + "loss": 0.9125, + "step": 6460 + }, + { + "epoch": 0.7768893164191667, + "grad_norm": 1.827303710081319, + "learning_rate": 4.996851386512331e-07, + "loss": 1.0004, + "step": 6461 + }, + { + "epoch": 0.7770095593098058, + "grad_norm": 1.854289754432393, + "learning_rate": 4.991701502355444e-07, + "loss": 1.0674, + "step": 6462 + }, + { + "epoch": 0.7771298022004449, + "grad_norm": 1.6103798405169534, + "learning_rate": 4.986553894945518e-07, + "loss": 0.9965, + "step": 6463 + }, + { + "epoch": 0.777250045091084, + "grad_norm": 1.9726270012528413, + "learning_rate": 4.981408565063416e-07, + "loss": 1.0968, + "step": 6464 + }, + { + "epoch": 0.777370287981723, + "grad_norm": 1.924156261673136, + "learning_rate": 4.976265513489701e-07, + "loss": 0.9983, + "step": 6465 + }, + { + "epoch": 0.7774905308723622, + "grad_norm": 2.294265614278849, + "learning_rate": 4.971124741004562e-07, + "loss": 1.0423, + "step": 6466 + }, + { + "epoch": 0.7776107737630013, + "grad_norm": 2.251235622633331, + "learning_rate": 4.965986248387846e-07, + "loss": 1.0026, + "step": 6467 + }, + { + "epoch": 0.7777310166536403, + "grad_norm": 4.88300937103116, + "learning_rate": 4.960850036419073e-07, + "loss": 1.0114, + "step": 6468 + }, + { + "epoch": 0.7778512595442795, + "grad_norm": 1.8018251649420125, + "learning_rate": 4.955716105877378e-07, + "loss": 1.0192, + "step": 6469 + }, + { + "epoch": 0.7779715024349185, + "grad_norm": 1.8189436714380416, + "learning_rate": 4.950584457541598e-07, + "loss": 1.067, + "step": 6470 + }, + { + "epoch": 0.7780917453255576, + "grad_norm": 1.4964462940585561, + "learning_rate": 4.945455092190183e-07, + "loss": 1.0557, + "step": 6471 + }, + { + "epoch": 0.7782119882161967, + "grad_norm": 0.6956912520015036, + "learning_rate": 4.940328010601271e-07, + "loss": 0.7975, + "step": 6472 + }, + { + "epoch": 0.7783322311068358, + "grad_norm": 1.7965405360213136, + "learning_rate": 4.935203213552621e-07, + "loss": 1.0083, + "step": 6473 + }, + { + "epoch": 0.7784524739974749, + "grad_norm": 2.728743590741881, + "learning_rate": 4.930080701821662e-07, + "loss": 0.9059, + "step": 6474 + }, + { + "epoch": 0.778572716888114, + "grad_norm": 2.108680441206064, + "learning_rate": 4.92496047618548e-07, + "loss": 1.008, + "step": 6475 + }, + { + "epoch": 0.7786929597787531, + "grad_norm": 1.6606781980516667, + "learning_rate": 4.919842537420811e-07, + "loss": 1.014, + "step": 6476 + }, + { + "epoch": 0.7788132026693921, + "grad_norm": 1.5723961059260083, + "learning_rate": 4.91472688630404e-07, + "loss": 1.0285, + "step": 6477 + }, + { + "epoch": 0.7789334455600313, + "grad_norm": 1.7721830796543314, + "learning_rate": 4.909613523611202e-07, + "loss": 0.9746, + "step": 6478 + }, + { + "epoch": 0.7790536884506704, + "grad_norm": 2.0478283940406747, + "learning_rate": 4.904502450117991e-07, + "loss": 0.9804, + "step": 6479 + }, + { + "epoch": 0.7791739313413094, + "grad_norm": 2.3059825297584693, + "learning_rate": 4.899393666599762e-07, + "loss": 0.9579, + "step": 6480 + }, + { + "epoch": 0.7792941742319486, + "grad_norm": 2.3557026558755605, + "learning_rate": 4.894287173831506e-07, + "loss": 0.9615, + "step": 6481 + }, + { + "epoch": 0.7794144171225876, + "grad_norm": 2.3638599538420575, + "learning_rate": 4.889182972587877e-07, + "loss": 1.0832, + "step": 6482 + }, + { + "epoch": 0.7795346600132267, + "grad_norm": 2.5172918259012538, + "learning_rate": 4.884081063643177e-07, + "loss": 0.8985, + "step": 6483 + }, + { + "epoch": 0.7796549029038659, + "grad_norm": 0.849746657932552, + "learning_rate": 4.878981447771353e-07, + "loss": 0.7627, + "step": 6484 + }, + { + "epoch": 0.7797751457945049, + "grad_norm": 2.434576384180656, + "learning_rate": 4.873884125746035e-07, + "loss": 0.9675, + "step": 6485 + }, + { + "epoch": 0.779895388685144, + "grad_norm": 2.2086901346501455, + "learning_rate": 4.868789098340456e-07, + "loss": 0.9661, + "step": 6486 + }, + { + "epoch": 0.7800156315757831, + "grad_norm": 2.4332866934221404, + "learning_rate": 4.863696366327543e-07, + "loss": 0.9655, + "step": 6487 + }, + { + "epoch": 0.7801358744664222, + "grad_norm": 1.8133491631592955, + "learning_rate": 4.85860593047986e-07, + "loss": 1.0173, + "step": 6488 + }, + { + "epoch": 0.7802561173570612, + "grad_norm": 1.6171235150196752, + "learning_rate": 4.853517791569613e-07, + "loss": 0.9859, + "step": 6489 + }, + { + "epoch": 0.7803763602477004, + "grad_norm": 1.9108291408106308, + "learning_rate": 4.848431950368684e-07, + "loss": 0.8985, + "step": 6490 + }, + { + "epoch": 0.7804966031383395, + "grad_norm": 0.7136681066454288, + "learning_rate": 4.843348407648569e-07, + "loss": 0.7942, + "step": 6491 + }, + { + "epoch": 0.7806168460289785, + "grad_norm": 2.578712165321983, + "learning_rate": 4.838267164180457e-07, + "loss": 1.068, + "step": 6492 + }, + { + "epoch": 0.7807370889196176, + "grad_norm": 1.9683080510198898, + "learning_rate": 4.833188220735156e-07, + "loss": 1.0774, + "step": 6493 + }, + { + "epoch": 0.7808573318102567, + "grad_norm": 2.855164309448364, + "learning_rate": 4.828111578083152e-07, + "loss": 0.9874, + "step": 6494 + }, + { + "epoch": 0.7809775747008958, + "grad_norm": 2.0691061859980864, + "learning_rate": 4.823037236994556e-07, + "loss": 1.0501, + "step": 6495 + }, + { + "epoch": 0.7810978175915348, + "grad_norm": 0.732016855788414, + "learning_rate": 4.817965198239136e-07, + "loss": 0.8007, + "step": 6496 + }, + { + "epoch": 0.781218060482174, + "grad_norm": 2.02494349459708, + "learning_rate": 4.812895462586331e-07, + "loss": 0.9807, + "step": 6497 + }, + { + "epoch": 0.7813383033728131, + "grad_norm": 2.0049647459892936, + "learning_rate": 4.807828030805207e-07, + "loss": 1.0548, + "step": 6498 + }, + { + "epoch": 0.7814585462634521, + "grad_norm": 1.9684770182924185, + "learning_rate": 4.802762903664495e-07, + "loss": 0.9114, + "step": 6499 + }, + { + "epoch": 0.7815787891540913, + "grad_norm": 2.51186264836145, + "learning_rate": 4.797700081932565e-07, + "loss": 0.9785, + "step": 6500 + }, + { + "epoch": 0.7816990320447303, + "grad_norm": 2.083573505544989, + "learning_rate": 4.792639566377442e-07, + "loss": 1.0516, + "step": 6501 + }, + { + "epoch": 0.7818192749353694, + "grad_norm": 1.9315013931747338, + "learning_rate": 4.78758135776681e-07, + "loss": 1.0085, + "step": 6502 + }, + { + "epoch": 0.7819395178260086, + "grad_norm": 3.1464577027888763, + "learning_rate": 4.782525456867989e-07, + "loss": 1.0257, + "step": 6503 + }, + { + "epoch": 0.7820597607166476, + "grad_norm": 1.6207988424807271, + "learning_rate": 4.777471864447959e-07, + "loss": 1.0681, + "step": 6504 + }, + { + "epoch": 0.7821800036072867, + "grad_norm": 2.124373848609803, + "learning_rate": 4.772420581273344e-07, + "loss": 1.0423, + "step": 6505 + }, + { + "epoch": 0.7823002464979258, + "grad_norm": 1.9736360741484058, + "learning_rate": 4.7673716081104134e-07, + "loss": 1.0008, + "step": 6506 + }, + { + "epoch": 0.7824204893885649, + "grad_norm": 1.8183909740222308, + "learning_rate": 4.762324945725109e-07, + "loss": 1.0804, + "step": 6507 + }, + { + "epoch": 0.782540732279204, + "grad_norm": 1.7617494703096903, + "learning_rate": 4.7572805948829844e-07, + "loss": 0.9936, + "step": 6508 + }, + { + "epoch": 0.7826609751698431, + "grad_norm": 1.783157264479076, + "learning_rate": 4.7522385563492795e-07, + "loss": 0.946, + "step": 6509 + }, + { + "epoch": 0.7827812180604822, + "grad_norm": 2.253401727917504, + "learning_rate": 4.747198830888863e-07, + "loss": 0.9393, + "step": 6510 + }, + { + "epoch": 0.7829014609511212, + "grad_norm": 2.3255435021218656, + "learning_rate": 4.742161419266251e-07, + "loss": 0.919, + "step": 6511 + }, + { + "epoch": 0.7830217038417604, + "grad_norm": 3.116581831676749, + "learning_rate": 4.7371263222456304e-07, + "loss": 0.8892, + "step": 6512 + }, + { + "epoch": 0.7831419467323995, + "grad_norm": 1.184753705677581, + "learning_rate": 4.7320935405908004e-07, + "loss": 0.8507, + "step": 6513 + }, + { + "epoch": 0.7832621896230385, + "grad_norm": 2.4311150285378154, + "learning_rate": 4.7270630750652475e-07, + "loss": 1.0781, + "step": 6514 + }, + { + "epoch": 0.7833824325136777, + "grad_norm": 1.6829128791531605, + "learning_rate": 4.7220349264320746e-07, + "loss": 1.0377, + "step": 6515 + }, + { + "epoch": 0.7835026754043167, + "grad_norm": 0.7484818829654479, + "learning_rate": 4.71700909545407e-07, + "loss": 0.7866, + "step": 6516 + }, + { + "epoch": 0.7836229182949558, + "grad_norm": 2.098283556690407, + "learning_rate": 4.711985582893627e-07, + "loss": 1.0053, + "step": 6517 + }, + { + "epoch": 0.783743161185595, + "grad_norm": 1.8928781749862709, + "learning_rate": 4.706964389512811e-07, + "loss": 0.953, + "step": 6518 + }, + { + "epoch": 0.783863404076234, + "grad_norm": 2.1352058368595936, + "learning_rate": 4.701945516073345e-07, + "loss": 1.1135, + "step": 6519 + }, + { + "epoch": 0.7839836469668731, + "grad_norm": 1.7992703601426716, + "learning_rate": 4.696928963336577e-07, + "loss": 0.9885, + "step": 6520 + }, + { + "epoch": 0.7841038898575122, + "grad_norm": 0.8633852934819515, + "learning_rate": 4.6919147320635224e-07, + "loss": 0.8475, + "step": 6521 + }, + { + "epoch": 0.7842241327481513, + "grad_norm": 2.3361101884067734, + "learning_rate": 4.6869028230148286e-07, + "loss": 0.9705, + "step": 6522 + }, + { + "epoch": 0.7843443756387903, + "grad_norm": 2.287522354433655, + "learning_rate": 4.6818932369507957e-07, + "loss": 0.8325, + "step": 6523 + }, + { + "epoch": 0.7844646185294295, + "grad_norm": 2.0101074877034764, + "learning_rate": 4.676885974631386e-07, + "loss": 1.1295, + "step": 6524 + }, + { + "epoch": 0.7845848614200686, + "grad_norm": 2.1202262664103486, + "learning_rate": 4.67188103681619e-07, + "loss": 1.0508, + "step": 6525 + }, + { + "epoch": 0.7847051043107076, + "grad_norm": 2.140277026775453, + "learning_rate": 4.666878424264453e-07, + "loss": 0.9315, + "step": 6526 + }, + { + "epoch": 0.7848253472013467, + "grad_norm": 1.5998388924549631, + "learning_rate": 4.661878137735069e-07, + "loss": 0.9749, + "step": 6527 + }, + { + "epoch": 0.7849455900919858, + "grad_norm": 1.9227754297872528, + "learning_rate": 4.656880177986571e-07, + "loss": 0.9864, + "step": 6528 + }, + { + "epoch": 0.7850658329826249, + "grad_norm": 2.057519505757125, + "learning_rate": 4.6518845457771607e-07, + "loss": 1.0531, + "step": 6529 + }, + { + "epoch": 0.7851860758732639, + "grad_norm": 1.90259764933277, + "learning_rate": 4.646891241864652e-07, + "loss": 1.0263, + "step": 6530 + }, + { + "epoch": 0.7853063187639031, + "grad_norm": 2.3426140639178747, + "learning_rate": 4.6419002670065397e-07, + "loss": 0.9672, + "step": 6531 + }, + { + "epoch": 0.7854265616545422, + "grad_norm": 2.6432359416605498, + "learning_rate": 4.6369116219599445e-07, + "loss": 1.1061, + "step": 6532 + }, + { + "epoch": 0.7855468045451812, + "grad_norm": 1.768754733992533, + "learning_rate": 4.631925307481637e-07, + "loss": 1.0291, + "step": 6533 + }, + { + "epoch": 0.7856670474358204, + "grad_norm": 2.1424098683364283, + "learning_rate": 4.6269413243280533e-07, + "loss": 0.993, + "step": 6534 + }, + { + "epoch": 0.7857872903264594, + "grad_norm": 2.7194937203964784, + "learning_rate": 4.621959673255236e-07, + "loss": 0.9872, + "step": 6535 + }, + { + "epoch": 0.7859075332170985, + "grad_norm": 2.1190066246876085, + "learning_rate": 4.6169803550189135e-07, + "loss": 1.1375, + "step": 6536 + }, + { + "epoch": 0.7860277761077377, + "grad_norm": 1.9876702049838444, + "learning_rate": 4.6120033703744355e-07, + "loss": 1.0142, + "step": 6537 + }, + { + "epoch": 0.7861480189983767, + "grad_norm": 2.2554425007132055, + "learning_rate": 4.607028720076822e-07, + "loss": 1.021, + "step": 6538 + }, + { + "epoch": 0.7862682618890158, + "grad_norm": 2.014937645867941, + "learning_rate": 4.6020564048807074e-07, + "loss": 0.9766, + "step": 6539 + }, + { + "epoch": 0.7863885047796549, + "grad_norm": 2.271388857957874, + "learning_rate": 4.5970864255403883e-07, + "loss": 0.9537, + "step": 6540 + }, + { + "epoch": 0.786508747670294, + "grad_norm": 2.332435253426294, + "learning_rate": 4.59211878280982e-07, + "loss": 1.056, + "step": 6541 + }, + { + "epoch": 0.786628990560933, + "grad_norm": 2.3262423897792464, + "learning_rate": 4.587153477442578e-07, + "loss": 0.9361, + "step": 6542 + }, + { + "epoch": 0.7867492334515722, + "grad_norm": 2.3740812833604585, + "learning_rate": 4.582190510191899e-07, + "loss": 1.0476, + "step": 6543 + }, + { + "epoch": 0.7868694763422113, + "grad_norm": 2.118643734948084, + "learning_rate": 4.5772298818106625e-07, + "loss": 1.109, + "step": 6544 + }, + { + "epoch": 0.7869897192328503, + "grad_norm": 2.440891156599716, + "learning_rate": 4.572271593051384e-07, + "loss": 0.9587, + "step": 6545 + }, + { + "epoch": 0.7871099621234895, + "grad_norm": 2.0425793949604008, + "learning_rate": 4.567315644666245e-07, + "loss": 1.0182, + "step": 6546 + }, + { + "epoch": 0.7872302050141285, + "grad_norm": 2.104592786893707, + "learning_rate": 4.5623620374070507e-07, + "loss": 1.0839, + "step": 6547 + }, + { + "epoch": 0.7873504479047676, + "grad_norm": 0.7639559634459471, + "learning_rate": 4.557410772025263e-07, + "loss": 0.821, + "step": 6548 + }, + { + "epoch": 0.7874706907954068, + "grad_norm": 1.9843215935196035, + "learning_rate": 4.5524618492719803e-07, + "loss": 0.8996, + "step": 6549 + }, + { + "epoch": 0.7875909336860458, + "grad_norm": 1.7151146329376417, + "learning_rate": 4.54751526989795e-07, + "loss": 1.0254, + "step": 6550 + }, + { + "epoch": 0.7877111765766849, + "grad_norm": 1.9990001801972073, + "learning_rate": 4.5425710346535775e-07, + "loss": 1.0312, + "step": 6551 + }, + { + "epoch": 0.787831419467324, + "grad_norm": 2.409137133225207, + "learning_rate": 4.537629144288877e-07, + "loss": 1.0601, + "step": 6552 + }, + { + "epoch": 0.7879516623579631, + "grad_norm": 2.378069650655064, + "learning_rate": 4.5326895995535477e-07, + "loss": 0.9869, + "step": 6553 + }, + { + "epoch": 0.7880719052486022, + "grad_norm": 2.5709317745443134, + "learning_rate": 4.527752401196907e-07, + "loss": 1.0785, + "step": 6554 + }, + { + "epoch": 0.7881921481392413, + "grad_norm": 2.7659305037389625, + "learning_rate": 4.5228175499679254e-07, + "loss": 0.9028, + "step": 6555 + }, + { + "epoch": 0.7883123910298804, + "grad_norm": 0.8479186675839321, + "learning_rate": 4.5178850466152174e-07, + "loss": 0.7825, + "step": 6556 + }, + { + "epoch": 0.7884326339205194, + "grad_norm": 2.536548484959973, + "learning_rate": 4.512954891887031e-07, + "loss": 1.0557, + "step": 6557 + }, + { + "epoch": 0.7885528768111585, + "grad_norm": 2.3871937979824716, + "learning_rate": 4.5080270865312806e-07, + "loss": 1.0751, + "step": 6558 + }, + { + "epoch": 0.7886731197017977, + "grad_norm": 1.8680733211512395, + "learning_rate": 4.5031016312954985e-07, + "loss": 0.9475, + "step": 6559 + }, + { + "epoch": 0.7887933625924367, + "grad_norm": 6.585570265191423, + "learning_rate": 4.498178526926886e-07, + "loss": 0.9801, + "step": 6560 + }, + { + "epoch": 0.7889136054830758, + "grad_norm": 2.1897892553576104, + "learning_rate": 4.4932577741722635e-07, + "loss": 0.9609, + "step": 6561 + }, + { + "epoch": 0.7890338483737149, + "grad_norm": 1.695765773105389, + "learning_rate": 4.4883393737780985e-07, + "loss": 0.9806, + "step": 6562 + }, + { + "epoch": 0.789154091264354, + "grad_norm": 2.4502063080475183, + "learning_rate": 4.4834233264905254e-07, + "loss": 1.0214, + "step": 6563 + }, + { + "epoch": 0.789274334154993, + "grad_norm": 2.6617376749515245, + "learning_rate": 4.478509633055294e-07, + "loss": 0.9517, + "step": 6564 + }, + { + "epoch": 0.7893945770456322, + "grad_norm": 2.5831039901163257, + "learning_rate": 4.473598294217813e-07, + "loss": 1.036, + "step": 6565 + }, + { + "epoch": 0.7895148199362713, + "grad_norm": 2.1654573758249995, + "learning_rate": 4.468689310723124e-07, + "loss": 0.9551, + "step": 6566 + }, + { + "epoch": 0.7896350628269103, + "grad_norm": 1.8641265934863354, + "learning_rate": 4.463782683315913e-07, + "loss": 1.0263, + "step": 6567 + }, + { + "epoch": 0.7897553057175495, + "grad_norm": 1.692446487910598, + "learning_rate": 4.458878412740523e-07, + "loss": 0.9712, + "step": 6568 + }, + { + "epoch": 0.7898755486081885, + "grad_norm": 4.458394463140833, + "learning_rate": 4.453976499740919e-07, + "loss": 1.0163, + "step": 6569 + }, + { + "epoch": 0.7899957914988276, + "grad_norm": 2.290306612502453, + "learning_rate": 4.4490769450607215e-07, + "loss": 1.0186, + "step": 6570 + }, + { + "epoch": 0.7901160343894668, + "grad_norm": 1.8880080088914304, + "learning_rate": 4.4441797494431845e-07, + "loss": 0.9581, + "step": 6571 + }, + { + "epoch": 0.7902362772801058, + "grad_norm": 2.0916884792249335, + "learning_rate": 4.439284913631207e-07, + "loss": 1.0154, + "step": 6572 + }, + { + "epoch": 0.7903565201707449, + "grad_norm": 2.6095688799253236, + "learning_rate": 4.434392438367347e-07, + "loss": 1.0767, + "step": 6573 + }, + { + "epoch": 0.790476763061384, + "grad_norm": 1.899204115063902, + "learning_rate": 4.4295023243937677e-07, + "loss": 0.979, + "step": 6574 + }, + { + "epoch": 0.7905970059520231, + "grad_norm": 1.8069600250813025, + "learning_rate": 4.4246145724523123e-07, + "loss": 1.0405, + "step": 6575 + }, + { + "epoch": 0.7907172488426621, + "grad_norm": 2.30416408996254, + "learning_rate": 4.41972918328444e-07, + "loss": 1.0057, + "step": 6576 + }, + { + "epoch": 0.7908374917333013, + "grad_norm": 2.092308459157717, + "learning_rate": 4.4148461576312646e-07, + "loss": 1.0094, + "step": 6577 + }, + { + "epoch": 0.7909577346239404, + "grad_norm": 2.1958118020244055, + "learning_rate": 4.4099654962335343e-07, + "loss": 0.9844, + "step": 6578 + }, + { + "epoch": 0.7910779775145794, + "grad_norm": 1.8745306699191804, + "learning_rate": 4.405087199831636e-07, + "loss": 0.9862, + "step": 6579 + }, + { + "epoch": 0.7911982204052186, + "grad_norm": 2.042176072804499, + "learning_rate": 4.400211269165619e-07, + "loss": 0.9048, + "step": 6580 + }, + { + "epoch": 0.7913184632958576, + "grad_norm": 1.5723378040497358, + "learning_rate": 4.3953377049751416e-07, + "loss": 1.0071, + "step": 6581 + }, + { + "epoch": 0.7914387061864967, + "grad_norm": 4.894390180279501, + "learning_rate": 4.390466507999537e-07, + "loss": 1.0162, + "step": 6582 + }, + { + "epoch": 0.7915589490771359, + "grad_norm": 2.192296681910574, + "learning_rate": 4.385597678977748e-07, + "loss": 0.9946, + "step": 6583 + }, + { + "epoch": 0.7916791919677749, + "grad_norm": 1.6212691981959069, + "learning_rate": 4.3807312186483726e-07, + "loss": 0.9915, + "step": 6584 + }, + { + "epoch": 0.791799434858414, + "grad_norm": 2.1620432112166417, + "learning_rate": 4.375867127749655e-07, + "loss": 1.0224, + "step": 6585 + }, + { + "epoch": 0.7919196777490531, + "grad_norm": 2.028020903703449, + "learning_rate": 4.3710054070194744e-07, + "loss": 0.9044, + "step": 6586 + }, + { + "epoch": 0.7920399206396922, + "grad_norm": 3.020173590515238, + "learning_rate": 4.3661460571953455e-07, + "loss": 0.902, + "step": 6587 + }, + { + "epoch": 0.7921601635303313, + "grad_norm": 1.6658845337859485, + "learning_rate": 4.36128907901443e-07, + "loss": 0.92, + "step": 6588 + }, + { + "epoch": 0.7922804064209703, + "grad_norm": 2.578968072084937, + "learning_rate": 4.356434473213519e-07, + "loss": 0.9671, + "step": 6589 + }, + { + "epoch": 0.7924006493116095, + "grad_norm": 1.663425233986704, + "learning_rate": 4.351582240529068e-07, + "loss": 1.0324, + "step": 6590 + }, + { + "epoch": 0.7925208922022485, + "grad_norm": 2.1808876370568773, + "learning_rate": 4.346732381697149e-07, + "loss": 0.8195, + "step": 6591 + }, + { + "epoch": 0.7926411350928876, + "grad_norm": 1.9296588355031952, + "learning_rate": 4.3418848974534825e-07, + "loss": 1.0519, + "step": 6592 + }, + { + "epoch": 0.7927613779835267, + "grad_norm": 1.6773236341473978, + "learning_rate": 4.3370397885334276e-07, + "loss": 0.9245, + "step": 6593 + }, + { + "epoch": 0.7928816208741658, + "grad_norm": 1.8591580304410167, + "learning_rate": 4.3321970556719777e-07, + "loss": 0.9922, + "step": 6594 + }, + { + "epoch": 0.7930018637648049, + "grad_norm": 2.4614665113763374, + "learning_rate": 4.3273566996037856e-07, + "loss": 0.955, + "step": 6595 + }, + { + "epoch": 0.793122106655444, + "grad_norm": 2.48938758963759, + "learning_rate": 4.322518721063113e-07, + "loss": 1.0407, + "step": 6596 + }, + { + "epoch": 0.7932423495460831, + "grad_norm": 2.250876785681813, + "learning_rate": 4.3176831207838906e-07, + "loss": 0.938, + "step": 6597 + }, + { + "epoch": 0.7933625924367221, + "grad_norm": 1.8548448658154852, + "learning_rate": 4.3128498994996685e-07, + "loss": 0.9806, + "step": 6598 + }, + { + "epoch": 0.7934828353273613, + "grad_norm": 1.9771409224981895, + "learning_rate": 4.308019057943646e-07, + "loss": 0.9512, + "step": 6599 + }, + { + "epoch": 0.7936030782180004, + "grad_norm": 1.7231424558113033, + "learning_rate": 4.3031905968486535e-07, + "loss": 0.9829, + "step": 6600 + }, + { + "epoch": 0.7937233211086394, + "grad_norm": 2.17411820086316, + "learning_rate": 4.298364516947162e-07, + "loss": 0.9252, + "step": 6601 + }, + { + "epoch": 0.7938435639992786, + "grad_norm": 2.017767780027581, + "learning_rate": 4.293540818971295e-07, + "loss": 0.8915, + "step": 6602 + }, + { + "epoch": 0.7939638068899176, + "grad_norm": 2.6561432760901464, + "learning_rate": 4.2887195036527934e-07, + "loss": 1.0055, + "step": 6603 + }, + { + "epoch": 0.7940840497805567, + "grad_norm": 2.5081741214887883, + "learning_rate": 4.28390057172306e-07, + "loss": 0.9695, + "step": 6604 + }, + { + "epoch": 0.7942042926711959, + "grad_norm": 2.913391690604492, + "learning_rate": 4.279084023913111e-07, + "loss": 0.9566, + "step": 6605 + }, + { + "epoch": 0.7943245355618349, + "grad_norm": 1.8390350832390427, + "learning_rate": 4.2742698609536096e-07, + "loss": 0.9313, + "step": 6606 + }, + { + "epoch": 0.794444778452474, + "grad_norm": 1.7981303598206704, + "learning_rate": 4.2694580835748706e-07, + "loss": 1.0248, + "step": 6607 + }, + { + "epoch": 0.7945650213431131, + "grad_norm": 2.8183005809929926, + "learning_rate": 4.264648692506836e-07, + "loss": 0.9784, + "step": 6608 + }, + { + "epoch": 0.7946852642337522, + "grad_norm": 2.1160930887669256, + "learning_rate": 4.2598416884790824e-07, + "loss": 0.9601, + "step": 6609 + }, + { + "epoch": 0.7948055071243912, + "grad_norm": 2.737875218565094, + "learning_rate": 4.255037072220828e-07, + "loss": 1.0431, + "step": 6610 + }, + { + "epoch": 0.7949257500150304, + "grad_norm": 1.7852198215909902, + "learning_rate": 4.2502348444609293e-07, + "loss": 0.9551, + "step": 6611 + }, + { + "epoch": 0.7950459929056695, + "grad_norm": 2.0556648743689014, + "learning_rate": 4.2454350059278844e-07, + "loss": 0.9281, + "step": 6612 + }, + { + "epoch": 0.7951662357963085, + "grad_norm": 2.5591283807485765, + "learning_rate": 4.240637557349824e-07, + "loss": 1.0822, + "step": 6613 + }, + { + "epoch": 0.7952864786869477, + "grad_norm": 3.5437730845619506, + "learning_rate": 4.235842499454516e-07, + "loss": 0.9049, + "step": 6614 + }, + { + "epoch": 0.7954067215775867, + "grad_norm": 1.7006201566594252, + "learning_rate": 4.2310498329693687e-07, + "loss": 1.0676, + "step": 6615 + }, + { + "epoch": 0.7955269644682258, + "grad_norm": 2.164205388058268, + "learning_rate": 4.2262595586214164e-07, + "loss": 1.0467, + "step": 6616 + }, + { + "epoch": 0.795647207358865, + "grad_norm": 1.759031153088023, + "learning_rate": 4.221471677137358e-07, + "loss": 1.0059, + "step": 6617 + }, + { + "epoch": 0.795767450249504, + "grad_norm": 1.5418624581768967, + "learning_rate": 4.216686189243492e-07, + "loss": 0.9359, + "step": 6618 + }, + { + "epoch": 0.7958876931401431, + "grad_norm": 2.307650765631972, + "learning_rate": 4.211903095665785e-07, + "loss": 0.9619, + "step": 6619 + }, + { + "epoch": 0.7960079360307821, + "grad_norm": 1.8515276040438546, + "learning_rate": 4.2071223971298277e-07, + "loss": 0.9924, + "step": 6620 + }, + { + "epoch": 0.7961281789214213, + "grad_norm": 2.0027249369688134, + "learning_rate": 4.2023440943608433e-07, + "loss": 0.8483, + "step": 6621 + }, + { + "epoch": 0.7962484218120603, + "grad_norm": 1.6500175821928111, + "learning_rate": 4.1975681880837023e-07, + "loss": 1.0174, + "step": 6622 + }, + { + "epoch": 0.7963686647026994, + "grad_norm": 1.9366126335712919, + "learning_rate": 4.192794679022895e-07, + "loss": 1.0616, + "step": 6623 + }, + { + "epoch": 0.7964889075933386, + "grad_norm": 1.9785931676523207, + "learning_rate": 4.1880235679025743e-07, + "loss": 0.9561, + "step": 6624 + }, + { + "epoch": 0.7966091504839776, + "grad_norm": 1.861542408187815, + "learning_rate": 4.1832548554464986e-07, + "loss": 0.8734, + "step": 6625 + }, + { + "epoch": 0.7967293933746167, + "grad_norm": 0.7534875448244224, + "learning_rate": 4.178488542378098e-07, + "loss": 0.8249, + "step": 6626 + }, + { + "epoch": 0.7968496362652558, + "grad_norm": 1.8094701106187232, + "learning_rate": 4.173724629420401e-07, + "loss": 1.1297, + "step": 6627 + }, + { + "epoch": 0.7969698791558949, + "grad_norm": 3.276695651149003, + "learning_rate": 4.168963117296087e-07, + "loss": 0.9198, + "step": 6628 + }, + { + "epoch": 0.797090122046534, + "grad_norm": 2.127658751467282, + "learning_rate": 4.1642040067274876e-07, + "loss": 0.9955, + "step": 6629 + }, + { + "epoch": 0.7972103649371731, + "grad_norm": 1.7818185668486637, + "learning_rate": 4.1594472984365493e-07, + "loss": 0.9609, + "step": 6630 + }, + { + "epoch": 0.7973306078278122, + "grad_norm": 2.463386019950705, + "learning_rate": 4.154692993144862e-07, + "loss": 1.0142, + "step": 6631 + }, + { + "epoch": 0.7974508507184512, + "grad_norm": 2.2237773407686117, + "learning_rate": 4.1499410915736476e-07, + "loss": 0.9557, + "step": 6632 + }, + { + "epoch": 0.7975710936090904, + "grad_norm": 0.7748459216743402, + "learning_rate": 4.145191594443762e-07, + "loss": 0.8794, + "step": 6633 + }, + { + "epoch": 0.7976913364997295, + "grad_norm": 1.6640379997921313, + "learning_rate": 4.140444502475713e-07, + "loss": 0.944, + "step": 6634 + }, + { + "epoch": 0.7978115793903685, + "grad_norm": 2.3789140672978086, + "learning_rate": 4.1356998163896216e-07, + "loss": 0.9356, + "step": 6635 + }, + { + "epoch": 0.7979318222810077, + "grad_norm": 2.186906461573879, + "learning_rate": 4.130957536905255e-07, + "loss": 0.9884, + "step": 6636 + }, + { + "epoch": 0.7980520651716467, + "grad_norm": 2.728983943345418, + "learning_rate": 4.1262176647420134e-07, + "loss": 0.9524, + "step": 6637 + }, + { + "epoch": 0.7981723080622858, + "grad_norm": 1.789138059394789, + "learning_rate": 4.121480200618923e-07, + "loss": 1.0356, + "step": 6638 + }, + { + "epoch": 0.798292550952925, + "grad_norm": 1.7864298544820452, + "learning_rate": 4.116745145254674e-07, + "loss": 1.0402, + "step": 6639 + }, + { + "epoch": 0.798412793843564, + "grad_norm": 0.777507132224583, + "learning_rate": 4.1120124993675476e-07, + "loss": 0.8171, + "step": 6640 + }, + { + "epoch": 0.7985330367342031, + "grad_norm": 2.312287295070106, + "learning_rate": 4.107282263675498e-07, + "loss": 0.8538, + "step": 6641 + }, + { + "epoch": 0.7986532796248422, + "grad_norm": 0.7778280465214946, + "learning_rate": 4.1025544388960907e-07, + "loss": 0.7623, + "step": 6642 + }, + { + "epoch": 0.7987735225154813, + "grad_norm": 3.0497221335488764, + "learning_rate": 4.097829025746538e-07, + "loss": 0.9555, + "step": 6643 + }, + { + "epoch": 0.7988937654061203, + "grad_norm": 0.6673309895504135, + "learning_rate": 4.0931060249436757e-07, + "loss": 0.8476, + "step": 6644 + }, + { + "epoch": 0.7990140082967595, + "grad_norm": 2.5953959562420787, + "learning_rate": 4.088385437203978e-07, + "loss": 0.9308, + "step": 6645 + }, + { + "epoch": 0.7991342511873986, + "grad_norm": 2.2157372450273667, + "learning_rate": 4.083667263243564e-07, + "loss": 1.0127, + "step": 6646 + }, + { + "epoch": 0.7992544940780376, + "grad_norm": 1.70513587320772, + "learning_rate": 4.0789515037781653e-07, + "loss": 0.954, + "step": 6647 + }, + { + "epoch": 0.7993747369686768, + "grad_norm": 1.8576651653518967, + "learning_rate": 4.0742381595231755e-07, + "loss": 1.0625, + "step": 6648 + }, + { + "epoch": 0.7994949798593158, + "grad_norm": 1.807511073562659, + "learning_rate": 4.06952723119359e-07, + "loss": 1.0197, + "step": 6649 + }, + { + "epoch": 0.7996152227499549, + "grad_norm": 1.9433648957600602, + "learning_rate": 4.0648187195040504e-07, + "loss": 0.9088, + "step": 6650 + }, + { + "epoch": 0.799735465640594, + "grad_norm": 0.8151082210419626, + "learning_rate": 4.060112625168848e-07, + "loss": 0.9135, + "step": 6651 + }, + { + "epoch": 0.7998557085312331, + "grad_norm": 2.033341724607736, + "learning_rate": 4.055408948901886e-07, + "loss": 0.9794, + "step": 6652 + }, + { + "epoch": 0.7999759514218722, + "grad_norm": 2.0501874372809477, + "learning_rate": 4.050707691416708e-07, + "loss": 0.9492, + "step": 6653 + }, + { + "epoch": 0.8000961943125112, + "grad_norm": 0.6800948928089319, + "learning_rate": 4.046008853426495e-07, + "loss": 0.8353, + "step": 6654 + }, + { + "epoch": 0.8002164372031504, + "grad_norm": 1.6622236954387273, + "learning_rate": 4.0413124356440464e-07, + "loss": 0.866, + "step": 6655 + }, + { + "epoch": 0.8003366800937894, + "grad_norm": 1.8993966775995852, + "learning_rate": 4.0366184387818223e-07, + "loss": 1.059, + "step": 6656 + }, + { + "epoch": 0.8004569229844285, + "grad_norm": 1.9102597822905638, + "learning_rate": 4.0319268635518797e-07, + "loss": 1.0885, + "step": 6657 + }, + { + "epoch": 0.8005771658750677, + "grad_norm": 1.5601728275127036, + "learning_rate": 4.027237710665943e-07, + "loss": 0.9892, + "step": 6658 + }, + { + "epoch": 0.8006974087657067, + "grad_norm": 1.8899681905057708, + "learning_rate": 4.022550980835344e-07, + "loss": 0.9346, + "step": 6659 + }, + { + "epoch": 0.8008176516563458, + "grad_norm": 2.201881869606132, + "learning_rate": 4.017866674771051e-07, + "loss": 1.0337, + "step": 6660 + }, + { + "epoch": 0.8009378945469849, + "grad_norm": 1.785201257852004, + "learning_rate": 4.013184793183688e-07, + "loss": 0.9816, + "step": 6661 + }, + { + "epoch": 0.801058137437624, + "grad_norm": 1.9775308038986503, + "learning_rate": 4.008505336783472e-07, + "loss": 0.962, + "step": 6662 + }, + { + "epoch": 0.801178380328263, + "grad_norm": 1.8576787055037478, + "learning_rate": 4.003828306280284e-07, + "loss": 1.0416, + "step": 6663 + }, + { + "epoch": 0.8012986232189022, + "grad_norm": 1.7384727801127613, + "learning_rate": 3.999153702383626e-07, + "loss": 1.0173, + "step": 6664 + }, + { + "epoch": 0.8014188661095413, + "grad_norm": 2.292589718426503, + "learning_rate": 3.9944815258026263e-07, + "loss": 0.973, + "step": 6665 + }, + { + "epoch": 0.8015391090001803, + "grad_norm": 1.8649994227840563, + "learning_rate": 3.989811777246057e-07, + "loss": 1.0697, + "step": 6666 + }, + { + "epoch": 0.8016593518908195, + "grad_norm": 0.8568461925135511, + "learning_rate": 3.985144457422305e-07, + "loss": 0.8997, + "step": 6667 + }, + { + "epoch": 0.8017795947814585, + "grad_norm": 1.9784310898039035, + "learning_rate": 3.9804795670394096e-07, + "loss": 1.0028, + "step": 6668 + }, + { + "epoch": 0.8018998376720976, + "grad_norm": 1.676864167419091, + "learning_rate": 3.975817106805022e-07, + "loss": 0.9406, + "step": 6669 + }, + { + "epoch": 0.8020200805627368, + "grad_norm": 1.8631030623287497, + "learning_rate": 3.97115707742645e-07, + "loss": 0.8831, + "step": 6670 + }, + { + "epoch": 0.8021403234533758, + "grad_norm": 2.4443953381528427, + "learning_rate": 3.966499479610599e-07, + "loss": 0.8909, + "step": 6671 + }, + { + "epoch": 0.8022605663440149, + "grad_norm": 2.1233020057143785, + "learning_rate": 3.9618443140640225e-07, + "loss": 0.8899, + "step": 6672 + }, + { + "epoch": 0.802380809234654, + "grad_norm": 0.6914248275416213, + "learning_rate": 3.957191581492918e-07, + "loss": 0.7503, + "step": 6673 + }, + { + "epoch": 0.8025010521252931, + "grad_norm": 2.5384573347328074, + "learning_rate": 3.952541282603097e-07, + "loss": 0.9488, + "step": 6674 + }, + { + "epoch": 0.8026212950159322, + "grad_norm": 2.5589714878009913, + "learning_rate": 3.9478934181000013e-07, + "loss": 1.0712, + "step": 6675 + }, + { + "epoch": 0.8027415379065713, + "grad_norm": 2.1536694993944803, + "learning_rate": 3.943247988688714e-07, + "loss": 1.0755, + "step": 6676 + }, + { + "epoch": 0.8028617807972104, + "grad_norm": 1.8047698031067039, + "learning_rate": 3.938604995073933e-07, + "loss": 0.9565, + "step": 6677 + }, + { + "epoch": 0.8029820236878494, + "grad_norm": 1.9983832261721575, + "learning_rate": 3.9339644379600157e-07, + "loss": 0.8884, + "step": 6678 + }, + { + "epoch": 0.8031022665784886, + "grad_norm": 2.2677169087051725, + "learning_rate": 3.929326318050907e-07, + "loss": 0.9493, + "step": 6679 + }, + { + "epoch": 0.8032225094691277, + "grad_norm": 2.0931004897488243, + "learning_rate": 3.924690636050225e-07, + "loss": 1.0278, + "step": 6680 + }, + { + "epoch": 0.8033427523597667, + "grad_norm": 2.014521451970775, + "learning_rate": 3.9200573926611915e-07, + "loss": 0.9629, + "step": 6681 + }, + { + "epoch": 0.8034629952504058, + "grad_norm": 2.212828026716937, + "learning_rate": 3.9154265885866613e-07, + "loss": 0.967, + "step": 6682 + }, + { + "epoch": 0.8035832381410449, + "grad_norm": 3.0197076388183657, + "learning_rate": 3.9107982245291394e-07, + "loss": 0.9862, + "step": 6683 + }, + { + "epoch": 0.803703481031684, + "grad_norm": 1.9097961835314325, + "learning_rate": 3.9061723011907245e-07, + "loss": 1.0114, + "step": 6684 + }, + { + "epoch": 0.803823723922323, + "grad_norm": 1.8022378917704858, + "learning_rate": 3.901548819273179e-07, + "loss": 1.0315, + "step": 6685 + }, + { + "epoch": 0.8039439668129622, + "grad_norm": 1.9289395508299958, + "learning_rate": 3.896927779477881e-07, + "loss": 0.9279, + "step": 6686 + }, + { + "epoch": 0.8040642097036013, + "grad_norm": 2.0455826035304003, + "learning_rate": 3.892309182505833e-07, + "loss": 0.907, + "step": 6687 + }, + { + "epoch": 0.8041844525942403, + "grad_norm": 2.2338954170766443, + "learning_rate": 3.887693029057675e-07, + "loss": 1.0945, + "step": 6688 + }, + { + "epoch": 0.8043046954848795, + "grad_norm": 1.720720860436409, + "learning_rate": 3.8830793198336684e-07, + "loss": 1.0493, + "step": 6689 + }, + { + "epoch": 0.8044249383755185, + "grad_norm": 1.74104168765199, + "learning_rate": 3.878468055533721e-07, + "loss": 0.9417, + "step": 6690 + }, + { + "epoch": 0.8045451812661576, + "grad_norm": 2.963474927708308, + "learning_rate": 3.8738592368573464e-07, + "loss": 1.0847, + "step": 6691 + }, + { + "epoch": 0.8046654241567968, + "grad_norm": 1.9682644442078314, + "learning_rate": 3.8692528645037137e-07, + "loss": 1.1199, + "step": 6692 + }, + { + "epoch": 0.8047856670474358, + "grad_norm": 2.4379424647238315, + "learning_rate": 3.8646489391715907e-07, + "loss": 1.0144, + "step": 6693 + }, + { + "epoch": 0.8049059099380749, + "grad_norm": 3.311573114812804, + "learning_rate": 3.8600474615593903e-07, + "loss": 1.1176, + "step": 6694 + }, + { + "epoch": 0.805026152828714, + "grad_norm": 0.7937414979854821, + "learning_rate": 3.8554484323651605e-07, + "loss": 0.8338, + "step": 6695 + }, + { + "epoch": 0.8051463957193531, + "grad_norm": 1.7997733582757185, + "learning_rate": 3.85085185228657e-07, + "loss": 1.026, + "step": 6696 + }, + { + "epoch": 0.8052666386099921, + "grad_norm": 1.890886903387558, + "learning_rate": 3.8462577220209114e-07, + "loss": 0.9664, + "step": 6697 + }, + { + "epoch": 0.8053868815006313, + "grad_norm": 0.7190161088006396, + "learning_rate": 3.8416660422651127e-07, + "loss": 0.8269, + "step": 6698 + }, + { + "epoch": 0.8055071243912704, + "grad_norm": 1.8237741978316597, + "learning_rate": 3.837076813715723e-07, + "loss": 0.9136, + "step": 6699 + }, + { + "epoch": 0.8056273672819094, + "grad_norm": 1.9517934914462527, + "learning_rate": 3.832490037068941e-07, + "loss": 0.9922, + "step": 6700 + }, + { + "epoch": 0.8057476101725486, + "grad_norm": 3.1838221555814927, + "learning_rate": 3.827905713020554e-07, + "loss": 0.9987, + "step": 6701 + }, + { + "epoch": 0.8058678530631876, + "grad_norm": 2.05195441704402, + "learning_rate": 3.823323842266017e-07, + "loss": 0.9243, + "step": 6702 + }, + { + "epoch": 0.8059880959538267, + "grad_norm": 2.318590957657401, + "learning_rate": 3.818744425500393e-07, + "loss": 0.9686, + "step": 6703 + }, + { + "epoch": 0.8061083388444659, + "grad_norm": 2.085592634288861, + "learning_rate": 3.8141674634183675e-07, + "loss": 1.042, + "step": 6704 + }, + { + "epoch": 0.8062285817351049, + "grad_norm": 1.9777698071398955, + "learning_rate": 3.809592956714278e-07, + "loss": 0.8964, + "step": 6705 + }, + { + "epoch": 0.806348824625744, + "grad_norm": 2.098428537856653, + "learning_rate": 3.805020906082057e-07, + "loss": 0.9869, + "step": 6706 + }, + { + "epoch": 0.8064690675163831, + "grad_norm": 2.212617808508356, + "learning_rate": 3.8004513122152917e-07, + "loss": 1.0478, + "step": 6707 + }, + { + "epoch": 0.8065893104070222, + "grad_norm": 2.0830812937549035, + "learning_rate": 3.79588417580718e-07, + "loss": 0.9072, + "step": 6708 + }, + { + "epoch": 0.8067095532976613, + "grad_norm": 2.2294880168940954, + "learning_rate": 3.791319497550558e-07, + "loss": 1.0017, + "step": 6709 + }, + { + "epoch": 0.8068297961883004, + "grad_norm": 1.936615403569845, + "learning_rate": 3.78675727813788e-07, + "loss": 0.9419, + "step": 6710 + }, + { + "epoch": 0.8069500390789395, + "grad_norm": 1.6614017616454584, + "learning_rate": 3.782197518261225e-07, + "loss": 0.9711, + "step": 6711 + }, + { + "epoch": 0.8070702819695785, + "grad_norm": 2.3423560511309494, + "learning_rate": 3.777640218612319e-07, + "loss": 1.1949, + "step": 6712 + }, + { + "epoch": 0.8071905248602176, + "grad_norm": 2.156688700842973, + "learning_rate": 3.773085379882488e-07, + "loss": 0.952, + "step": 6713 + }, + { + "epoch": 0.8073107677508568, + "grad_norm": 2.0898450049280695, + "learning_rate": 3.768533002762715e-07, + "loss": 0.9969, + "step": 6714 + }, + { + "epoch": 0.8074310106414958, + "grad_norm": 1.890458028687106, + "learning_rate": 3.763983087943572e-07, + "loss": 1.006, + "step": 6715 + }, + { + "epoch": 0.8075512535321349, + "grad_norm": 1.6393282170370362, + "learning_rate": 3.759435636115282e-07, + "loss": 1.0434, + "step": 6716 + }, + { + "epoch": 0.807671496422774, + "grad_norm": 2.118422032181762, + "learning_rate": 3.7548906479676967e-07, + "loss": 0.9669, + "step": 6717 + }, + { + "epoch": 0.8077917393134131, + "grad_norm": 1.8774793127472007, + "learning_rate": 3.7503481241902855e-07, + "loss": 0.9519, + "step": 6718 + }, + { + "epoch": 0.8079119822040521, + "grad_norm": 2.275284374195778, + "learning_rate": 3.745808065472145e-07, + "loss": 1.0399, + "step": 6719 + }, + { + "epoch": 0.8080322250946913, + "grad_norm": 1.7041256266209226, + "learning_rate": 3.741270472501994e-07, + "loss": 1.0018, + "step": 6720 + }, + { + "epoch": 0.8081524679853304, + "grad_norm": 1.6646589902538622, + "learning_rate": 3.736735345968183e-07, + "loss": 0.9656, + "step": 6721 + }, + { + "epoch": 0.8082727108759694, + "grad_norm": 1.745289388997087, + "learning_rate": 3.7322026865586986e-07, + "loss": 1.0244, + "step": 6722 + }, + { + "epoch": 0.8083929537666086, + "grad_norm": 2.099493433390389, + "learning_rate": 3.7276724949611206e-07, + "loss": 0.9711, + "step": 6723 + }, + { + "epoch": 0.8085131966572476, + "grad_norm": 5.564184555261862, + "learning_rate": 3.723144771862694e-07, + "loss": 0.9919, + "step": 6724 + }, + { + "epoch": 0.8086334395478867, + "grad_norm": 1.613532464740176, + "learning_rate": 3.718619517950263e-07, + "loss": 1.0023, + "step": 6725 + }, + { + "epoch": 0.8087536824385259, + "grad_norm": 1.8809765613845775, + "learning_rate": 3.714096733910301e-07, + "loss": 1.0053, + "step": 6726 + }, + { + "epoch": 0.8088739253291649, + "grad_norm": 2.606655325348646, + "learning_rate": 3.709576420428926e-07, + "loss": 0.9392, + "step": 6727 + }, + { + "epoch": 0.808994168219804, + "grad_norm": 2.1047019907021345, + "learning_rate": 3.7050585781918463e-07, + "loss": 0.9727, + "step": 6728 + }, + { + "epoch": 0.8091144111104431, + "grad_norm": 5.455294551851526, + "learning_rate": 3.700543207884428e-07, + "loss": 0.9274, + "step": 6729 + }, + { + "epoch": 0.8092346540010822, + "grad_norm": 1.67924806036304, + "learning_rate": 3.6960303101916466e-07, + "loss": 0.9464, + "step": 6730 + }, + { + "epoch": 0.8093548968917212, + "grad_norm": 0.7448188314396833, + "learning_rate": 3.6915198857981047e-07, + "loss": 0.7923, + "step": 6731 + }, + { + "epoch": 0.8094751397823604, + "grad_norm": 1.6759815585412365, + "learning_rate": 3.687011935388027e-07, + "loss": 0.9186, + "step": 6732 + }, + { + "epoch": 0.8095953826729995, + "grad_norm": 1.9967062172875876, + "learning_rate": 3.6825064596452646e-07, + "loss": 0.9651, + "step": 6733 + }, + { + "epoch": 0.8097156255636385, + "grad_norm": 1.7067788502707928, + "learning_rate": 3.678003459253305e-07, + "loss": 0.941, + "step": 6734 + }, + { + "epoch": 0.8098358684542777, + "grad_norm": 2.255549580493565, + "learning_rate": 3.673502934895236e-07, + "loss": 0.9762, + "step": 6735 + }, + { + "epoch": 0.8099561113449167, + "grad_norm": 0.7041205669489037, + "learning_rate": 3.669004887253802e-07, + "loss": 0.8169, + "step": 6736 + }, + { + "epoch": 0.8100763542355558, + "grad_norm": 1.611782162593488, + "learning_rate": 3.664509317011335e-07, + "loss": 1.0279, + "step": 6737 + }, + { + "epoch": 0.810196597126195, + "grad_norm": 1.9309702945362333, + "learning_rate": 3.6600162248498134e-07, + "loss": 0.9747, + "step": 6738 + }, + { + "epoch": 0.810316840016834, + "grad_norm": 1.88485265343116, + "learning_rate": 3.6555256114508426e-07, + "loss": 0.9984, + "step": 6739 + }, + { + "epoch": 0.8104370829074731, + "grad_norm": 2.104246786373658, + "learning_rate": 3.651037477495642e-07, + "loss": 0.9648, + "step": 6740 + }, + { + "epoch": 0.8105573257981122, + "grad_norm": 2.064910289487558, + "learning_rate": 3.6465518236650584e-07, + "loss": 0.9164, + "step": 6741 + }, + { + "epoch": 0.8106775686887513, + "grad_norm": 1.7782932972833023, + "learning_rate": 3.642068650639558e-07, + "loss": 1.0196, + "step": 6742 + }, + { + "epoch": 0.8107978115793903, + "grad_norm": 2.884642477641886, + "learning_rate": 3.6375879590992334e-07, + "loss": 0.8883, + "step": 6743 + }, + { + "epoch": 0.8109180544700295, + "grad_norm": 1.965111655443368, + "learning_rate": 3.6331097497238173e-07, + "loss": 1.0498, + "step": 6744 + }, + { + "epoch": 0.8110382973606686, + "grad_norm": 2.008076691223035, + "learning_rate": 3.628634023192627e-07, + "loss": 1.0373, + "step": 6745 + }, + { + "epoch": 0.8111585402513076, + "grad_norm": 2.20857830008469, + "learning_rate": 3.624160780184644e-07, + "loss": 0.992, + "step": 6746 + }, + { + "epoch": 0.8112787831419467, + "grad_norm": 1.8134887892106004, + "learning_rate": 3.6196900213784496e-07, + "loss": 0.978, + "step": 6747 + }, + { + "epoch": 0.8113990260325858, + "grad_norm": 1.990923192372942, + "learning_rate": 3.6152217474522527e-07, + "loss": 1.1028, + "step": 6748 + }, + { + "epoch": 0.8115192689232249, + "grad_norm": 2.1462717688583477, + "learning_rate": 3.6107559590838975e-07, + "loss": 0.9623, + "step": 6749 + }, + { + "epoch": 0.811639511813864, + "grad_norm": 2.5158675177586045, + "learning_rate": 3.606292656950822e-07, + "loss": 0.8976, + "step": 6750 + }, + { + "epoch": 0.8117597547045031, + "grad_norm": 1.8726228267788714, + "learning_rate": 3.601831841730121e-07, + "loss": 1.0999, + "step": 6751 + }, + { + "epoch": 0.8118799975951422, + "grad_norm": 1.9330225245253263, + "learning_rate": 3.5973735140984916e-07, + "loss": 0.9642, + "step": 6752 + }, + { + "epoch": 0.8120002404857812, + "grad_norm": 2.167836778077796, + "learning_rate": 3.5929176747322607e-07, + "loss": 1.0292, + "step": 6753 + }, + { + "epoch": 0.8121204833764204, + "grad_norm": 0.81519720909457, + "learning_rate": 3.588464324307372e-07, + "loss": 0.7985, + "step": 6754 + }, + { + "epoch": 0.8122407262670595, + "grad_norm": 1.9989605228901757, + "learning_rate": 3.584013463499391e-07, + "loss": 0.9906, + "step": 6755 + }, + { + "epoch": 0.8123609691576985, + "grad_norm": 0.7448339561425086, + "learning_rate": 3.579565092983521e-07, + "loss": 0.882, + "step": 6756 + }, + { + "epoch": 0.8124812120483377, + "grad_norm": 6.02096740502983, + "learning_rate": 3.575119213434565e-07, + "loss": 1.0764, + "step": 6757 + }, + { + "epoch": 0.8126014549389767, + "grad_norm": 1.961412288741868, + "learning_rate": 3.5706758255269765e-07, + "loss": 1.0541, + "step": 6758 + }, + { + "epoch": 0.8127216978296158, + "grad_norm": 1.6276339678261311, + "learning_rate": 3.566234929934795e-07, + "loss": 0.9328, + "step": 6759 + }, + { + "epoch": 0.812841940720255, + "grad_norm": 1.8169010267831014, + "learning_rate": 3.561796527331706e-07, + "loss": 0.9533, + "step": 6760 + }, + { + "epoch": 0.812962183610894, + "grad_norm": 1.81022066636402, + "learning_rate": 3.5573606183910163e-07, + "loss": 1.0138, + "step": 6761 + }, + { + "epoch": 0.8130824265015331, + "grad_norm": 1.724208763657115, + "learning_rate": 3.5529272037856493e-07, + "loss": 1.0219, + "step": 6762 + }, + { + "epoch": 0.8132026693921722, + "grad_norm": 0.715342946669066, + "learning_rate": 3.548496284188149e-07, + "loss": 0.776, + "step": 6763 + }, + { + "epoch": 0.8133229122828113, + "grad_norm": 1.9669793052884064, + "learning_rate": 3.544067860270681e-07, + "loss": 1.0286, + "step": 6764 + }, + { + "epoch": 0.8134431551734503, + "grad_norm": 1.9357231975967621, + "learning_rate": 3.539641932705029e-07, + "loss": 0.9499, + "step": 6765 + }, + { + "epoch": 0.8135633980640895, + "grad_norm": 2.6989196523301477, + "learning_rate": 3.53521850216262e-07, + "loss": 1.0101, + "step": 6766 + }, + { + "epoch": 0.8136836409547286, + "grad_norm": 4.511216912701856, + "learning_rate": 3.530797569314461e-07, + "loss": 1.0049, + "step": 6767 + }, + { + "epoch": 0.8138038838453676, + "grad_norm": 1.9038208664887188, + "learning_rate": 3.5263791348312235e-07, + "loss": 1.0158, + "step": 6768 + }, + { + "epoch": 0.8139241267360068, + "grad_norm": 1.7359312481428544, + "learning_rate": 3.521963199383171e-07, + "loss": 0.9442, + "step": 6769 + }, + { + "epoch": 0.8140443696266458, + "grad_norm": 2.971213994688684, + "learning_rate": 3.517549763640197e-07, + "loss": 1.003, + "step": 6770 + }, + { + "epoch": 0.8141646125172849, + "grad_norm": 2.0569112949188635, + "learning_rate": 3.513138828271829e-07, + "loss": 0.949, + "step": 6771 + }, + { + "epoch": 0.8142848554079241, + "grad_norm": 1.9789502343411214, + "learning_rate": 3.508730393947179e-07, + "loss": 0.9405, + "step": 6772 + }, + { + "epoch": 0.8144050982985631, + "grad_norm": 2.175851598181073, + "learning_rate": 3.504324461335024e-07, + "loss": 0.9557, + "step": 6773 + }, + { + "epoch": 0.8145253411892022, + "grad_norm": 1.6778700821406658, + "learning_rate": 3.499921031103732e-07, + "loss": 1.1188, + "step": 6774 + }, + { + "epoch": 0.8146455840798413, + "grad_norm": 3.2464496587283955, + "learning_rate": 3.4955201039212987e-07, + "loss": 1.0199, + "step": 6775 + }, + { + "epoch": 0.8147658269704804, + "grad_norm": 2.1820287675088816, + "learning_rate": 3.4911216804553465e-07, + "loss": 0.8887, + "step": 6776 + }, + { + "epoch": 0.8148860698611194, + "grad_norm": 2.115770266798571, + "learning_rate": 3.4867257613731017e-07, + "loss": 0.9401, + "step": 6777 + }, + { + "epoch": 0.8150063127517585, + "grad_norm": 1.7404818546189829, + "learning_rate": 3.4823323473414343e-07, + "loss": 1.0926, + "step": 6778 + }, + { + "epoch": 0.8151265556423977, + "grad_norm": 1.8469342682134289, + "learning_rate": 3.477941439026812e-07, + "loss": 0.9951, + "step": 6779 + }, + { + "epoch": 0.8152467985330367, + "grad_norm": 1.872167227579107, + "learning_rate": 3.473553037095349e-07, + "loss": 0.9683, + "step": 6780 + }, + { + "epoch": 0.8153670414236758, + "grad_norm": 2.820852127651185, + "learning_rate": 3.469167142212743e-07, + "loss": 1.0704, + "step": 6781 + }, + { + "epoch": 0.8154872843143149, + "grad_norm": 2.2724860123979176, + "learning_rate": 3.4647837550443337e-07, + "loss": 0.8669, + "step": 6782 + }, + { + "epoch": 0.815607527204954, + "grad_norm": 6.589346187962388, + "learning_rate": 3.460402876255086e-07, + "loss": 0.9854, + "step": 6783 + }, + { + "epoch": 0.815727770095593, + "grad_norm": 2.1494126205058204, + "learning_rate": 3.456024506509574e-07, + "loss": 0.9564, + "step": 6784 + }, + { + "epoch": 0.8158480129862322, + "grad_norm": 1.7262478364769307, + "learning_rate": 3.4516486464719873e-07, + "loss": 0.9771, + "step": 6785 + }, + { + "epoch": 0.8159682558768713, + "grad_norm": 1.7269395787592867, + "learning_rate": 3.4472752968061445e-07, + "loss": 0.8574, + "step": 6786 + }, + { + "epoch": 0.8160884987675103, + "grad_norm": 1.952000469727538, + "learning_rate": 3.442904458175475e-07, + "loss": 0.9739, + "step": 6787 + }, + { + "epoch": 0.8162087416581495, + "grad_norm": 1.6917605128197686, + "learning_rate": 3.438536131243044e-07, + "loss": 0.9952, + "step": 6788 + }, + { + "epoch": 0.8163289845487885, + "grad_norm": 2.126164509951303, + "learning_rate": 3.434170316671503e-07, + "loss": 0.8562, + "step": 6789 + }, + { + "epoch": 0.8164492274394276, + "grad_norm": 2.5884365211382723, + "learning_rate": 3.4298070151231583e-07, + "loss": 1.1283, + "step": 6790 + }, + { + "epoch": 0.8165694703300668, + "grad_norm": 2.0115978612039225, + "learning_rate": 3.425446227259916e-07, + "loss": 0.8367, + "step": 6791 + }, + { + "epoch": 0.8166897132207058, + "grad_norm": 1.9009823794243106, + "learning_rate": 3.421087953743296e-07, + "loss": 1.0605, + "step": 6792 + }, + { + "epoch": 0.8168099561113449, + "grad_norm": 1.8941437315094267, + "learning_rate": 3.416732195234464e-07, + "loss": 1.0351, + "step": 6793 + }, + { + "epoch": 0.816930199001984, + "grad_norm": 1.4803262204127685, + "learning_rate": 3.4123789523941613e-07, + "loss": 1.0313, + "step": 6794 + }, + { + "epoch": 0.8170504418926231, + "grad_norm": 1.590458706480125, + "learning_rate": 3.4080282258827884e-07, + "loss": 0.8732, + "step": 6795 + }, + { + "epoch": 0.8171706847832622, + "grad_norm": 2.6070101866587105, + "learning_rate": 3.403680016360342e-07, + "loss": 0.9601, + "step": 6796 + }, + { + "epoch": 0.8172909276739013, + "grad_norm": 1.5621009317042542, + "learning_rate": 3.3993343244864403e-07, + "loss": 0.9155, + "step": 6797 + }, + { + "epoch": 0.8174111705645404, + "grad_norm": 1.8282039005826765, + "learning_rate": 3.394991150920323e-07, + "loss": 0.9656, + "step": 6798 + }, + { + "epoch": 0.8175314134551794, + "grad_norm": 2.49952655123859, + "learning_rate": 3.3906504963208396e-07, + "loss": 0.9823, + "step": 6799 + }, + { + "epoch": 0.8176516563458186, + "grad_norm": 2.0397625302494333, + "learning_rate": 3.3863123613464774e-07, + "loss": 0.902, + "step": 6800 + }, + { + "epoch": 0.8177718992364577, + "grad_norm": 1.8026499950627428, + "learning_rate": 3.381976746655317e-07, + "loss": 0.9843, + "step": 6801 + }, + { + "epoch": 0.8178921421270967, + "grad_norm": 1.9291400827947696, + "learning_rate": 3.3776436529050756e-07, + "loss": 0.9075, + "step": 6802 + }, + { + "epoch": 0.8180123850177359, + "grad_norm": 1.6255970371632347, + "learning_rate": 3.373313080753073e-07, + "loss": 0.9625, + "step": 6803 + }, + { + "epoch": 0.8181326279083749, + "grad_norm": 1.5734920670291483, + "learning_rate": 3.3689850308562527e-07, + "loss": 1.0138, + "step": 6804 + }, + { + "epoch": 0.818252870799014, + "grad_norm": 1.8269459109095596, + "learning_rate": 3.364659503871183e-07, + "loss": 1.013, + "step": 6805 + }, + { + "epoch": 0.8183731136896532, + "grad_norm": 1.9828777521938319, + "learning_rate": 3.3603365004540417e-07, + "loss": 1.0711, + "step": 6806 + }, + { + "epoch": 0.8184933565802922, + "grad_norm": 3.2749014337281612, + "learning_rate": 3.356016021260624e-07, + "loss": 1.0093, + "step": 6807 + }, + { + "epoch": 0.8186135994709313, + "grad_norm": 4.117429317198889, + "learning_rate": 3.35169806694634e-07, + "loss": 0.8858, + "step": 6808 + }, + { + "epoch": 0.8187338423615703, + "grad_norm": 0.7283916214446913, + "learning_rate": 3.3473826381662186e-07, + "loss": 0.8437, + "step": 6809 + }, + { + "epoch": 0.8188540852522095, + "grad_norm": 2.0523446654109705, + "learning_rate": 3.3430697355749216e-07, + "loss": 1.0546, + "step": 6810 + }, + { + "epoch": 0.8189743281428485, + "grad_norm": 2.28874772026411, + "learning_rate": 3.3387593598266907e-07, + "loss": 0.9953, + "step": 6811 + }, + { + "epoch": 0.8190945710334876, + "grad_norm": 1.730733123630895, + "learning_rate": 3.3344515115754225e-07, + "loss": 1.0166, + "step": 6812 + }, + { + "epoch": 0.8192148139241268, + "grad_norm": 2.6792258707469014, + "learning_rate": 3.33014619147461e-07, + "loss": 1.0381, + "step": 6813 + }, + { + "epoch": 0.8193350568147658, + "grad_norm": 1.858230045966403, + "learning_rate": 3.325843400177362e-07, + "loss": 0.9491, + "step": 6814 + }, + { + "epoch": 0.8194552997054049, + "grad_norm": 2.1084999812711662, + "learning_rate": 3.32154313833642e-07, + "loss": 0.9738, + "step": 6815 + }, + { + "epoch": 0.819575542596044, + "grad_norm": 4.29468294583251, + "learning_rate": 3.3172454066041164e-07, + "loss": 0.8308, + "step": 6816 + }, + { + "epoch": 0.8196957854866831, + "grad_norm": 1.9931377463353352, + "learning_rate": 3.3129502056324234e-07, + "loss": 0.996, + "step": 6817 + }, + { + "epoch": 0.8198160283773221, + "grad_norm": 0.8042409824628264, + "learning_rate": 3.3086575360729165e-07, + "loss": 0.8354, + "step": 6818 + }, + { + "epoch": 0.8199362712679613, + "grad_norm": 1.853630681381044, + "learning_rate": 3.3043673985767906e-07, + "loss": 0.952, + "step": 6819 + }, + { + "epoch": 0.8200565141586004, + "grad_norm": 1.7130248692450374, + "learning_rate": 3.3000797937948564e-07, + "loss": 1.0111, + "step": 6820 + }, + { + "epoch": 0.8201767570492394, + "grad_norm": 0.9454529082635896, + "learning_rate": 3.295794722377534e-07, + "loss": 0.8874, + "step": 6821 + }, + { + "epoch": 0.8202969999398786, + "grad_norm": 1.7982421318146171, + "learning_rate": 3.291512184974876e-07, + "loss": 1.0361, + "step": 6822 + }, + { + "epoch": 0.8204172428305176, + "grad_norm": 1.8549052135588604, + "learning_rate": 3.2872321822365346e-07, + "loss": 0.8997, + "step": 6823 + }, + { + "epoch": 0.8205374857211567, + "grad_norm": 8.24228376721032, + "learning_rate": 3.282954714811783e-07, + "loss": 0.9659, + "step": 6824 + }, + { + "epoch": 0.8206577286117959, + "grad_norm": 2.304038884331485, + "learning_rate": 3.2786797833495093e-07, + "loss": 0.9509, + "step": 6825 + }, + { + "epoch": 0.8207779715024349, + "grad_norm": 2.262917736120878, + "learning_rate": 3.274407388498213e-07, + "loss": 0.967, + "step": 6826 + }, + { + "epoch": 0.820898214393074, + "grad_norm": 1.944561916920254, + "learning_rate": 3.270137530906021e-07, + "loss": 0.9785, + "step": 6827 + }, + { + "epoch": 0.8210184572837131, + "grad_norm": 2.667268764551768, + "learning_rate": 3.265870211220665e-07, + "loss": 1.0702, + "step": 6828 + }, + { + "epoch": 0.8211387001743522, + "grad_norm": 2.1942800897571595, + "learning_rate": 3.2616054300894934e-07, + "loss": 1.0509, + "step": 6829 + }, + { + "epoch": 0.8212589430649913, + "grad_norm": 2.7236922494579083, + "learning_rate": 3.2573431881594693e-07, + "loss": 1.0841, + "step": 6830 + }, + { + "epoch": 0.8213791859556304, + "grad_norm": 2.337399130700084, + "learning_rate": 3.2530834860771663e-07, + "loss": 0.8962, + "step": 6831 + }, + { + "epoch": 0.8214994288462695, + "grad_norm": 2.6182825331329935, + "learning_rate": 3.248826324488794e-07, + "loss": 0.9768, + "step": 6832 + }, + { + "epoch": 0.8216196717369085, + "grad_norm": 1.8024297686265889, + "learning_rate": 3.244571704040138e-07, + "loss": 1.1131, + "step": 6833 + }, + { + "epoch": 0.8217399146275477, + "grad_norm": 2.2806447937708327, + "learning_rate": 3.2403196253766374e-07, + "loss": 0.9743, + "step": 6834 + }, + { + "epoch": 0.8218601575181868, + "grad_norm": 2.72063906581195, + "learning_rate": 3.2360700891433254e-07, + "loss": 1.028, + "step": 6835 + }, + { + "epoch": 0.8219804004088258, + "grad_norm": 0.9463509144759505, + "learning_rate": 3.231823095984847e-07, + "loss": 0.8107, + "step": 6836 + }, + { + "epoch": 0.822100643299465, + "grad_norm": 2.0426358887235305, + "learning_rate": 3.2275786465454814e-07, + "loss": 0.9969, + "step": 6837 + }, + { + "epoch": 0.822220886190104, + "grad_norm": 1.8407146231264377, + "learning_rate": 3.2233367414690917e-07, + "loss": 0.9912, + "step": 6838 + }, + { + "epoch": 0.8223411290807431, + "grad_norm": 2.3187944474805757, + "learning_rate": 3.219097381399183e-07, + "loss": 1.0866, + "step": 6839 + }, + { + "epoch": 0.8224613719713821, + "grad_norm": 4.402743195332084, + "learning_rate": 3.2148605669788584e-07, + "loss": 1.0484, + "step": 6840 + }, + { + "epoch": 0.8225816148620213, + "grad_norm": 2.3950045492809946, + "learning_rate": 3.2106262988508405e-07, + "loss": 1.0095, + "step": 6841 + }, + { + "epoch": 0.8227018577526604, + "grad_norm": 2.6479797867985972, + "learning_rate": 3.206394577657465e-07, + "loss": 0.9767, + "step": 6842 + }, + { + "epoch": 0.8228221006432994, + "grad_norm": 6.533322530205547, + "learning_rate": 3.202165404040675e-07, + "loss": 0.9602, + "step": 6843 + }, + { + "epoch": 0.8229423435339386, + "grad_norm": 2.2368764679811473, + "learning_rate": 3.1979387786420396e-07, + "loss": 0.9834, + "step": 6844 + }, + { + "epoch": 0.8230625864245776, + "grad_norm": 2.0290314516400145, + "learning_rate": 3.1937147021027346e-07, + "loss": 1.057, + "step": 6845 + }, + { + "epoch": 0.8231828293152167, + "grad_norm": 14.390366807706869, + "learning_rate": 3.189493175063547e-07, + "loss": 0.9986, + "step": 6846 + }, + { + "epoch": 0.8233030722058559, + "grad_norm": 2.7103934937722167, + "learning_rate": 3.1852741981648776e-07, + "loss": 0.9103, + "step": 6847 + }, + { + "epoch": 0.8234233150964949, + "grad_norm": 2.0459715035804225, + "learning_rate": 3.1810577720467404e-07, + "loss": 0.9384, + "step": 6848 + }, + { + "epoch": 0.823543557987134, + "grad_norm": 1.8295435170400862, + "learning_rate": 3.176843897348769e-07, + "loss": 0.8028, + "step": 6849 + }, + { + "epoch": 0.8236638008777731, + "grad_norm": 3.3960777772010506, + "learning_rate": 3.1726325747102034e-07, + "loss": 0.9902, + "step": 6850 + }, + { + "epoch": 0.8237840437684122, + "grad_norm": 1.629072222041265, + "learning_rate": 3.1684238047698974e-07, + "loss": 0.8775, + "step": 6851 + }, + { + "epoch": 0.8239042866590512, + "grad_norm": 2.744767412709468, + "learning_rate": 3.1642175881663155e-07, + "loss": 0.768, + "step": 6852 + }, + { + "epoch": 0.8240245295496904, + "grad_norm": 2.4539558315763625, + "learning_rate": 3.160013925537537e-07, + "loss": 1.0757, + "step": 6853 + }, + { + "epoch": 0.8241447724403295, + "grad_norm": 2.3872307510742465, + "learning_rate": 3.155812817521266e-07, + "loss": 0.9914, + "step": 6854 + }, + { + "epoch": 0.8242650153309685, + "grad_norm": 2.1757149540716147, + "learning_rate": 3.151614264754787e-07, + "loss": 1.0163, + "step": 6855 + }, + { + "epoch": 0.8243852582216077, + "grad_norm": 2.481107374267148, + "learning_rate": 3.147418267875035e-07, + "loss": 1.0307, + "step": 6856 + }, + { + "epoch": 0.8245055011122467, + "grad_norm": 2.627260143391489, + "learning_rate": 3.1432248275185315e-07, + "loss": 0.894, + "step": 6857 + }, + { + "epoch": 0.8246257440028858, + "grad_norm": 2.5621181296541997, + "learning_rate": 3.139033944321412e-07, + "loss": 1.0079, + "step": 6858 + }, + { + "epoch": 0.824745986893525, + "grad_norm": 1.6480933820044557, + "learning_rate": 3.1348456189194507e-07, + "loss": 1.0261, + "step": 6859 + }, + { + "epoch": 0.824866229784164, + "grad_norm": 1.865945442833172, + "learning_rate": 3.1306598519479876e-07, + "loss": 1.0655, + "step": 6860 + }, + { + "epoch": 0.8249864726748031, + "grad_norm": 1.6059440410456307, + "learning_rate": 3.1264766440420177e-07, + "loss": 1.0173, + "step": 6861 + }, + { + "epoch": 0.8251067155654422, + "grad_norm": 1.9241174780416657, + "learning_rate": 3.122295995836124e-07, + "loss": 0.9296, + "step": 6862 + }, + { + "epoch": 0.8252269584560813, + "grad_norm": 2.7288498342707075, + "learning_rate": 3.118117907964508e-07, + "loss": 1.013, + "step": 6863 + }, + { + "epoch": 0.8253472013467203, + "grad_norm": 1.9108272692249608, + "learning_rate": 3.1139423810609856e-07, + "loss": 1.0444, + "step": 6864 + }, + { + "epoch": 0.8254674442373595, + "grad_norm": 1.9751481737011407, + "learning_rate": 3.1097694157589714e-07, + "loss": 0.9885, + "step": 6865 + }, + { + "epoch": 0.8255876871279986, + "grad_norm": 3.179796371658091, + "learning_rate": 3.105599012691511e-07, + "loss": 1.0011, + "step": 6866 + }, + { + "epoch": 0.8257079300186376, + "grad_norm": 1.4432728990555013, + "learning_rate": 3.101431172491249e-07, + "loss": 1.0584, + "step": 6867 + }, + { + "epoch": 0.8258281729092768, + "grad_norm": 2.647386370445649, + "learning_rate": 3.097265895790444e-07, + "loss": 0.9569, + "step": 6868 + }, + { + "epoch": 0.8259484157999158, + "grad_norm": 1.9401352096907354, + "learning_rate": 3.093103183220962e-07, + "loss": 1.0736, + "step": 6869 + }, + { + "epoch": 0.8260686586905549, + "grad_norm": 0.8280471909154669, + "learning_rate": 3.0889430354142796e-07, + "loss": 0.8304, + "step": 6870 + }, + { + "epoch": 0.826188901581194, + "grad_norm": 1.7867665446652194, + "learning_rate": 3.084785453001497e-07, + "loss": 0.9314, + "step": 6871 + }, + { + "epoch": 0.8263091444718331, + "grad_norm": 2.5476704858816746, + "learning_rate": 3.080630436613314e-07, + "loss": 1.0576, + "step": 6872 + }, + { + "epoch": 0.8264293873624722, + "grad_norm": 2.283016722602437, + "learning_rate": 3.076477986880039e-07, + "loss": 1.0956, + "step": 6873 + }, + { + "epoch": 0.8265496302531112, + "grad_norm": 2.6136093047780613, + "learning_rate": 3.0723281044315986e-07, + "loss": 0.9277, + "step": 6874 + }, + { + "epoch": 0.8266698731437504, + "grad_norm": 2.1233400705650833, + "learning_rate": 3.068180789897521e-07, + "loss": 1.0003, + "step": 6875 + }, + { + "epoch": 0.8267901160343895, + "grad_norm": 1.5525587066509454, + "learning_rate": 3.064036043906966e-07, + "loss": 1.0514, + "step": 6876 + }, + { + "epoch": 0.8269103589250285, + "grad_norm": 2.0812970955133916, + "learning_rate": 3.059893867088668e-07, + "loss": 0.9172, + "step": 6877 + }, + { + "epoch": 0.8270306018156677, + "grad_norm": 2.0364611171173297, + "learning_rate": 3.055754260071004e-07, + "loss": 0.9046, + "step": 6878 + }, + { + "epoch": 0.8271508447063067, + "grad_norm": 2.629255569172092, + "learning_rate": 3.051617223481948e-07, + "loss": 0.9715, + "step": 6879 + }, + { + "epoch": 0.8272710875969458, + "grad_norm": 2.854760788016978, + "learning_rate": 3.047482757949078e-07, + "loss": 0.99, + "step": 6880 + }, + { + "epoch": 0.827391330487585, + "grad_norm": 2.6535787277546694, + "learning_rate": 3.043350864099605e-07, + "loss": 1.094, + "step": 6881 + }, + { + "epoch": 0.827511573378224, + "grad_norm": 2.3035109798235127, + "learning_rate": 3.039221542560315e-07, + "loss": 1.0413, + "step": 6882 + }, + { + "epoch": 0.8276318162688631, + "grad_norm": 1.8616661253233657, + "learning_rate": 3.0350947939576356e-07, + "loss": 0.97, + "step": 6883 + }, + { + "epoch": 0.8277520591595022, + "grad_norm": 1.6139872855639341, + "learning_rate": 3.0309706189175876e-07, + "loss": 0.9574, + "step": 6884 + }, + { + "epoch": 0.8278723020501413, + "grad_norm": 0.7696769305365775, + "learning_rate": 3.0268490180658045e-07, + "loss": 0.8109, + "step": 6885 + }, + { + "epoch": 0.8279925449407803, + "grad_norm": 2.493447872971691, + "learning_rate": 3.0227299920275305e-07, + "loss": 1.0278, + "step": 6886 + }, + { + "epoch": 0.8281127878314195, + "grad_norm": 2.065952301641713, + "learning_rate": 3.018613541427613e-07, + "loss": 1.09, + "step": 6887 + }, + { + "epoch": 0.8282330307220586, + "grad_norm": 1.7935556020633625, + "learning_rate": 3.0144996668905243e-07, + "loss": 0.9723, + "step": 6888 + }, + { + "epoch": 0.8283532736126976, + "grad_norm": 2.86745491132067, + "learning_rate": 3.010388369040331e-07, + "loss": 1.0575, + "step": 6889 + }, + { + "epoch": 0.8284735165033368, + "grad_norm": 1.4758565516972657, + "learning_rate": 3.0062796485007156e-07, + "loss": 1.0619, + "step": 6890 + }, + { + "epoch": 0.8285937593939758, + "grad_norm": 3.453811646283058, + "learning_rate": 3.002173505894965e-07, + "loss": 0.8948, + "step": 6891 + }, + { + "epoch": 0.8287140022846149, + "grad_norm": 2.7694579035377616, + "learning_rate": 2.998069941845973e-07, + "loss": 0.8612, + "step": 6892 + }, + { + "epoch": 0.8288342451752541, + "grad_norm": 0.7215891614823147, + "learning_rate": 2.993968956976258e-07, + "loss": 0.813, + "step": 6893 + }, + { + "epoch": 0.8289544880658931, + "grad_norm": 1.963748452169584, + "learning_rate": 2.9898705519079313e-07, + "loss": 0.9375, + "step": 6894 + }, + { + "epoch": 0.8290747309565322, + "grad_norm": 1.9985869303319963, + "learning_rate": 2.985774727262715e-07, + "loss": 0.9877, + "step": 6895 + }, + { + "epoch": 0.8291949738471713, + "grad_norm": 2.7584303901246243, + "learning_rate": 2.981681483661949e-07, + "loss": 1.0508, + "step": 6896 + }, + { + "epoch": 0.8293152167378104, + "grad_norm": 1.6710695840123493, + "learning_rate": 2.9775908217265633e-07, + "loss": 0.9465, + "step": 6897 + }, + { + "epoch": 0.8294354596284494, + "grad_norm": 0.841626001703892, + "learning_rate": 2.9735027420771253e-07, + "loss": 0.7412, + "step": 6898 + }, + { + "epoch": 0.8295557025190886, + "grad_norm": 1.9395789406218265, + "learning_rate": 2.969417245333774e-07, + "loss": 0.9494, + "step": 6899 + }, + { + "epoch": 0.8296759454097277, + "grad_norm": 3.4744611583058704, + "learning_rate": 2.9653343321162915e-07, + "loss": 1.0138, + "step": 6900 + }, + { + "epoch": 0.8297961883003667, + "grad_norm": 1.9663580032631571, + "learning_rate": 2.9612540030440446e-07, + "loss": 0.8842, + "step": 6901 + }, + { + "epoch": 0.8299164311910058, + "grad_norm": 0.8767414814666005, + "learning_rate": 2.9571762587360206e-07, + "loss": 0.878, + "step": 6902 + }, + { + "epoch": 0.8300366740816449, + "grad_norm": 2.202058574593869, + "learning_rate": 2.953101099810806e-07, + "loss": 0.9779, + "step": 6903 + }, + { + "epoch": 0.830156916972284, + "grad_norm": 2.172003433491784, + "learning_rate": 2.9490285268865965e-07, + "loss": 1.061, + "step": 6904 + }, + { + "epoch": 0.830277159862923, + "grad_norm": 5.127719064810122, + "learning_rate": 2.9449585405812085e-07, + "loss": 1.0326, + "step": 6905 + }, + { + "epoch": 0.8303974027535622, + "grad_norm": 2.0077531264020356, + "learning_rate": 2.940891141512043e-07, + "loss": 0.974, + "step": 6906 + }, + { + "epoch": 0.8305176456442013, + "grad_norm": 2.1016538238536344, + "learning_rate": 2.9368263302961385e-07, + "loss": 0.9574, + "step": 6907 + }, + { + "epoch": 0.8306378885348403, + "grad_norm": 2.3056741962266716, + "learning_rate": 2.9327641075501075e-07, + "loss": 1.033, + "step": 6908 + }, + { + "epoch": 0.8307581314254795, + "grad_norm": 2.3547948376163057, + "learning_rate": 2.9287044738901866e-07, + "loss": 0.9002, + "step": 6909 + }, + { + "epoch": 0.8308783743161186, + "grad_norm": 3.1725344418699497, + "learning_rate": 2.9246474299322274e-07, + "loss": 1.1456, + "step": 6910 + }, + { + "epoch": 0.8309986172067576, + "grad_norm": 0.8958203884047042, + "learning_rate": 2.920592976291678e-07, + "loss": 0.869, + "step": 6911 + }, + { + "epoch": 0.8311188600973968, + "grad_norm": 2.279900975001008, + "learning_rate": 2.916541113583595e-07, + "loss": 1.0418, + "step": 6912 + }, + { + "epoch": 0.8312391029880358, + "grad_norm": 2.6599735863887153, + "learning_rate": 2.912491842422642e-07, + "loss": 0.9019, + "step": 6913 + }, + { + "epoch": 0.8313593458786749, + "grad_norm": 1.9524572833244744, + "learning_rate": 2.9084451634230857e-07, + "loss": 0.9443, + "step": 6914 + }, + { + "epoch": 0.831479588769314, + "grad_norm": 2.4088629865055315, + "learning_rate": 2.9044010771988125e-07, + "loss": 0.95, + "step": 6915 + }, + { + "epoch": 0.8315998316599531, + "grad_norm": 1.9912271615902002, + "learning_rate": 2.900359584363303e-07, + "loss": 0.9583, + "step": 6916 + }, + { + "epoch": 0.8317200745505922, + "grad_norm": 2.2421530478352705, + "learning_rate": 2.8963206855296494e-07, + "loss": 1.0838, + "step": 6917 + }, + { + "epoch": 0.8318403174412313, + "grad_norm": 1.993875603620806, + "learning_rate": 2.892284381310548e-07, + "loss": 1.0097, + "step": 6918 + }, + { + "epoch": 0.8319605603318704, + "grad_norm": 2.3758160042596295, + "learning_rate": 2.888250672318302e-07, + "loss": 0.9616, + "step": 6919 + }, + { + "epoch": 0.8320808032225094, + "grad_norm": 1.921503418767758, + "learning_rate": 2.884219559164831e-07, + "loss": 0.924, + "step": 6920 + }, + { + "epoch": 0.8322010461131486, + "grad_norm": 2.0212989611525667, + "learning_rate": 2.880191042461635e-07, + "loss": 1.0504, + "step": 6921 + }, + { + "epoch": 0.8323212890037877, + "grad_norm": 1.9923723322077866, + "learning_rate": 2.876165122819849e-07, + "loss": 1.0407, + "step": 6922 + }, + { + "epoch": 0.8324415318944267, + "grad_norm": 1.5763118639187275, + "learning_rate": 2.872141800850201e-07, + "loss": 1.0338, + "step": 6923 + }, + { + "epoch": 0.8325617747850659, + "grad_norm": 1.8211921891126863, + "learning_rate": 2.868121077163024e-07, + "loss": 0.9706, + "step": 6924 + }, + { + "epoch": 0.8326820176757049, + "grad_norm": 2.061066100162312, + "learning_rate": 2.864102952368257e-07, + "loss": 0.9614, + "step": 6925 + }, + { + "epoch": 0.832802260566344, + "grad_norm": 1.3922585311905005, + "learning_rate": 2.860087427075444e-07, + "loss": 0.831, + "step": 6926 + }, + { + "epoch": 0.8329225034569832, + "grad_norm": 2.6017641098213145, + "learning_rate": 2.856074501893744e-07, + "loss": 1.0998, + "step": 6927 + }, + { + "epoch": 0.8330427463476222, + "grad_norm": 1.8157287646793931, + "learning_rate": 2.8520641774319054e-07, + "loss": 1.0539, + "step": 6928 + }, + { + "epoch": 0.8331629892382613, + "grad_norm": 2.056048736462189, + "learning_rate": 2.848056454298309e-07, + "loss": 0.9968, + "step": 6929 + }, + { + "epoch": 0.8332832321289004, + "grad_norm": 2.104070705328131, + "learning_rate": 2.844051333100905e-07, + "loss": 0.8931, + "step": 6930 + }, + { + "epoch": 0.8334034750195395, + "grad_norm": 2.143112408102339, + "learning_rate": 2.840048814447269e-07, + "loss": 1.0806, + "step": 6931 + }, + { + "epoch": 0.8335237179101785, + "grad_norm": 2.515985498601851, + "learning_rate": 2.836048898944587e-07, + "loss": 0.9754, + "step": 6932 + }, + { + "epoch": 0.8336439608008177, + "grad_norm": 2.5443304725823133, + "learning_rate": 2.832051587199642e-07, + "loss": 0.9513, + "step": 6933 + }, + { + "epoch": 0.8337642036914568, + "grad_norm": 0.8155284342109298, + "learning_rate": 2.828056879818821e-07, + "loss": 0.8145, + "step": 6934 + }, + { + "epoch": 0.8338844465820958, + "grad_norm": 1.9140050139844826, + "learning_rate": 2.824064777408117e-07, + "loss": 1.0723, + "step": 6935 + }, + { + "epoch": 0.8340046894727349, + "grad_norm": 1.7194007335249946, + "learning_rate": 2.8200752805731263e-07, + "loss": 0.9965, + "step": 6936 + }, + { + "epoch": 0.834124932363374, + "grad_norm": 1.6359401844953574, + "learning_rate": 2.8160883899190625e-07, + "loss": 1.0467, + "step": 6937 + }, + { + "epoch": 0.8342451752540131, + "grad_norm": 2.4586969742441274, + "learning_rate": 2.8121041060507234e-07, + "loss": 0.9715, + "step": 6938 + }, + { + "epoch": 0.8343654181446521, + "grad_norm": 1.6870806844170294, + "learning_rate": 2.808122429572528e-07, + "loss": 0.9505, + "step": 6939 + }, + { + "epoch": 0.8344856610352913, + "grad_norm": 3.929048945152417, + "learning_rate": 2.804143361088489e-07, + "loss": 0.9928, + "step": 6940 + }, + { + "epoch": 0.8346059039259304, + "grad_norm": 2.7868157748738858, + "learning_rate": 2.8001669012022277e-07, + "loss": 1.0136, + "step": 6941 + }, + { + "epoch": 0.8347261468165694, + "grad_norm": 1.6651186747829236, + "learning_rate": 2.7961930505169795e-07, + "loss": 0.9352, + "step": 6942 + }, + { + "epoch": 0.8348463897072086, + "grad_norm": 2.3464955842264934, + "learning_rate": 2.792221809635558e-07, + "loss": 1.0011, + "step": 6943 + }, + { + "epoch": 0.8349666325978476, + "grad_norm": 2.056854729588746, + "learning_rate": 2.788253179160411e-07, + "loss": 0.9852, + "step": 6944 + }, + { + "epoch": 0.8350868754884867, + "grad_norm": 2.154850491782958, + "learning_rate": 2.7842871596935725e-07, + "loss": 0.8874, + "step": 6945 + }, + { + "epoch": 0.8352071183791259, + "grad_norm": 1.8508906694806377, + "learning_rate": 2.780323751836682e-07, + "loss": 0.9295, + "step": 6946 + }, + { + "epoch": 0.8353273612697649, + "grad_norm": 1.7995476790182745, + "learning_rate": 2.7763629561909876e-07, + "loss": 1.025, + "step": 6947 + }, + { + "epoch": 0.835447604160404, + "grad_norm": 2.2198060437265994, + "learning_rate": 2.772404773357335e-07, + "loss": 1.0037, + "step": 6948 + }, + { + "epoch": 0.8355678470510431, + "grad_norm": 2.1905981739703533, + "learning_rate": 2.7684492039361853e-07, + "loss": 1.0231, + "step": 6949 + }, + { + "epoch": 0.8356880899416822, + "grad_norm": 1.736278142735325, + "learning_rate": 2.764496248527586e-07, + "loss": 1.0744, + "step": 6950 + }, + { + "epoch": 0.8358083328323213, + "grad_norm": 2.2796206992011987, + "learning_rate": 2.760545907731211e-07, + "loss": 1.019, + "step": 6951 + }, + { + "epoch": 0.8359285757229604, + "grad_norm": 1.753821152149793, + "learning_rate": 2.75659818214631e-07, + "loss": 0.9139, + "step": 6952 + }, + { + "epoch": 0.8360488186135995, + "grad_norm": 1.7941251860821614, + "learning_rate": 2.752653072371749e-07, + "loss": 1.0203, + "step": 6953 + }, + { + "epoch": 0.8361690615042385, + "grad_norm": 1.8289107891745051, + "learning_rate": 2.7487105790060105e-07, + "loss": 0.9853, + "step": 6954 + }, + { + "epoch": 0.8362893043948777, + "grad_norm": 2.0106609637477213, + "learning_rate": 2.7447707026471587e-07, + "loss": 0.9314, + "step": 6955 + }, + { + "epoch": 0.8364095472855168, + "grad_norm": 3.193263509168834, + "learning_rate": 2.740833443892874e-07, + "loss": 1.0374, + "step": 6956 + }, + { + "epoch": 0.8365297901761558, + "grad_norm": 1.7539105319712138, + "learning_rate": 2.7368988033404327e-07, + "loss": 1.0375, + "step": 6957 + }, + { + "epoch": 0.836650033066795, + "grad_norm": 1.6550745841792651, + "learning_rate": 2.732966781586712e-07, + "loss": 1.0835, + "step": 6958 + }, + { + "epoch": 0.836770275957434, + "grad_norm": 1.6252917247956093, + "learning_rate": 2.729037379228205e-07, + "loss": 0.9039, + "step": 6959 + }, + { + "epoch": 0.8368905188480731, + "grad_norm": 1.4740984505237593, + "learning_rate": 2.725110596860998e-07, + "loss": 1.0426, + "step": 6960 + }, + { + "epoch": 0.8370107617387123, + "grad_norm": 3.5334185128171205, + "learning_rate": 2.7211864350807776e-07, + "loss": 0.9348, + "step": 6961 + }, + { + "epoch": 0.8371310046293513, + "grad_norm": 1.6452974903838524, + "learning_rate": 2.717264894482836e-07, + "loss": 0.9757, + "step": 6962 + }, + { + "epoch": 0.8372512475199904, + "grad_norm": 2.121585009257835, + "learning_rate": 2.7133459756620646e-07, + "loss": 1.0493, + "step": 6963 + }, + { + "epoch": 0.8373714904106295, + "grad_norm": 1.8048469506044553, + "learning_rate": 2.7094296792129733e-07, + "loss": 0.9757, + "step": 6964 + }, + { + "epoch": 0.8374917333012686, + "grad_norm": 1.8840387553624696, + "learning_rate": 2.7055160057296424e-07, + "loss": 0.9947, + "step": 6965 + }, + { + "epoch": 0.8376119761919076, + "grad_norm": 1.6704701733728065, + "learning_rate": 2.7016049558057896e-07, + "loss": 0.9638, + "step": 6966 + }, + { + "epoch": 0.8377322190825467, + "grad_norm": 1.7919601229675588, + "learning_rate": 2.6976965300347074e-07, + "loss": 0.9447, + "step": 6967 + }, + { + "epoch": 0.8378524619731859, + "grad_norm": 3.2424094250571915, + "learning_rate": 2.693790729009309e-07, + "loss": 0.9299, + "step": 6968 + }, + { + "epoch": 0.8379727048638249, + "grad_norm": 1.7863967558662623, + "learning_rate": 2.6898875533220946e-07, + "loss": 1.1194, + "step": 6969 + }, + { + "epoch": 0.838092947754464, + "grad_norm": 1.7897901096573299, + "learning_rate": 2.685987003565171e-07, + "loss": 1.0512, + "step": 6970 + }, + { + "epoch": 0.8382131906451031, + "grad_norm": 2.474184356537131, + "learning_rate": 2.6820890803302566e-07, + "loss": 0.9942, + "step": 6971 + }, + { + "epoch": 0.8383334335357422, + "grad_norm": 2.749355847603527, + "learning_rate": 2.6781937842086557e-07, + "loss": 1.0563, + "step": 6972 + }, + { + "epoch": 0.8384536764263812, + "grad_norm": 1.9134567859356182, + "learning_rate": 2.6743011157912933e-07, + "loss": 0.9092, + "step": 6973 + }, + { + "epoch": 0.8385739193170204, + "grad_norm": 2.001738031509006, + "learning_rate": 2.6704110756686725e-07, + "loss": 0.8902, + "step": 6974 + }, + { + "epoch": 0.8386941622076595, + "grad_norm": 2.1835047613309952, + "learning_rate": 2.6665236644309085e-07, + "loss": 1.0722, + "step": 6975 + }, + { + "epoch": 0.8388144050982985, + "grad_norm": 2.155510042907354, + "learning_rate": 2.662638882667727e-07, + "loss": 1.031, + "step": 6976 + }, + { + "epoch": 0.8389346479889377, + "grad_norm": 2.077450291375167, + "learning_rate": 2.658756730968443e-07, + "loss": 0.9657, + "step": 6977 + }, + { + "epoch": 0.8390548908795767, + "grad_norm": 2.1972734374861114, + "learning_rate": 2.654877209921975e-07, + "loss": 1.1223, + "step": 6978 + }, + { + "epoch": 0.8391751337702158, + "grad_norm": 3.021493054841686, + "learning_rate": 2.651000320116843e-07, + "loss": 0.8627, + "step": 6979 + }, + { + "epoch": 0.839295376660855, + "grad_norm": 2.1328518483091923, + "learning_rate": 2.647126062141163e-07, + "loss": 0.9971, + "step": 6980 + }, + { + "epoch": 0.839415619551494, + "grad_norm": 1.8412858045609755, + "learning_rate": 2.643254436582669e-07, + "loss": 1.0756, + "step": 6981 + }, + { + "epoch": 0.8395358624421331, + "grad_norm": 3.8534078452442, + "learning_rate": 2.6393854440286743e-07, + "loss": 1.0616, + "step": 6982 + }, + { + "epoch": 0.8396561053327722, + "grad_norm": 2.1090139574554025, + "learning_rate": 2.6355190850661045e-07, + "loss": 0.945, + "step": 6983 + }, + { + "epoch": 0.8397763482234113, + "grad_norm": 1.6957704830785674, + "learning_rate": 2.631655360281486e-07, + "loss": 1.1035, + "step": 6984 + }, + { + "epoch": 0.8398965911140504, + "grad_norm": 1.944460149731864, + "learning_rate": 2.6277942702609323e-07, + "loss": 0.8938, + "step": 6985 + }, + { + "epoch": 0.8400168340046895, + "grad_norm": 2.0163067987888255, + "learning_rate": 2.623935815590186e-07, + "loss": 1.112, + "step": 6986 + }, + { + "epoch": 0.8401370768953286, + "grad_norm": 2.0593971262829993, + "learning_rate": 2.6200799968545516e-07, + "loss": 1.0434, + "step": 6987 + }, + { + "epoch": 0.8402573197859676, + "grad_norm": 1.035196094825725, + "learning_rate": 2.616226814638969e-07, + "loss": 0.8019, + "step": 6988 + }, + { + "epoch": 0.8403775626766068, + "grad_norm": 3.2665447783274546, + "learning_rate": 2.612376269527954e-07, + "loss": 1.011, + "step": 6989 + }, + { + "epoch": 0.8404978055672458, + "grad_norm": 1.6938101444615796, + "learning_rate": 2.608528362105635e-07, + "loss": 0.9108, + "step": 6990 + }, + { + "epoch": 0.8406180484578849, + "grad_norm": 2.9720293103257824, + "learning_rate": 2.6046830929557374e-07, + "loss": 0.9675, + "step": 6991 + }, + { + "epoch": 0.8407382913485241, + "grad_norm": 1.9168382097960641, + "learning_rate": 2.6008404626615776e-07, + "loss": 1.0846, + "step": 6992 + }, + { + "epoch": 0.8408585342391631, + "grad_norm": 3.1316619053903074, + "learning_rate": 2.597000471806092e-07, + "loss": 0.9724, + "step": 6993 + }, + { + "epoch": 0.8409787771298022, + "grad_norm": 2.163238809320109, + "learning_rate": 2.593163120971793e-07, + "loss": 0.969, + "step": 6994 + }, + { + "epoch": 0.8410990200204413, + "grad_norm": 1.99931216332876, + "learning_rate": 2.5893284107408165e-07, + "loss": 0.9259, + "step": 6995 + }, + { + "epoch": 0.8412192629110804, + "grad_norm": 1.7793203241371272, + "learning_rate": 2.5854963416948726e-07, + "loss": 1.013, + "step": 6996 + }, + { + "epoch": 0.8413395058017195, + "grad_norm": 1.854005256513697, + "learning_rate": 2.5816669144152816e-07, + "loss": 0.9305, + "step": 6997 + }, + { + "epoch": 0.8414597486923585, + "grad_norm": 0.8653904621220196, + "learning_rate": 2.5778401294829777e-07, + "loss": 0.9007, + "step": 6998 + }, + { + "epoch": 0.8415799915829977, + "grad_norm": 1.7545247166333866, + "learning_rate": 2.574015987478473e-07, + "loss": 0.8864, + "step": 6999 + }, + { + "epoch": 0.8417002344736367, + "grad_norm": 2.2108240536541617, + "learning_rate": 2.570194488981887e-07, + "loss": 1.1074, + "step": 7000 + }, + { + "epoch": 0.8418204773642758, + "grad_norm": 0.8441798563344489, + "learning_rate": 2.566375634572939e-07, + "loss": 0.841, + "step": 7001 + }, + { + "epoch": 0.841940720254915, + "grad_norm": 1.8417646419418994, + "learning_rate": 2.562559424830943e-07, + "loss": 0.9984, + "step": 7002 + }, + { + "epoch": 0.842060963145554, + "grad_norm": 3.5670676145655444, + "learning_rate": 2.5587458603348256e-07, + "loss": 0.9371, + "step": 7003 + }, + { + "epoch": 0.8421812060361931, + "grad_norm": 2.2457251735632706, + "learning_rate": 2.554934941663085e-07, + "loss": 1.0735, + "step": 7004 + }, + { + "epoch": 0.8423014489268322, + "grad_norm": 2.0608495408637006, + "learning_rate": 2.5511266693938484e-07, + "loss": 0.9693, + "step": 7005 + }, + { + "epoch": 0.8424216918174713, + "grad_norm": 1.6350180591670815, + "learning_rate": 2.547321044104822e-07, + "loss": 1.0128, + "step": 7006 + }, + { + "epoch": 0.8425419347081103, + "grad_norm": 1.9464000434834718, + "learning_rate": 2.5435180663733113e-07, + "loss": 1.0023, + "step": 7007 + }, + { + "epoch": 0.8426621775987495, + "grad_norm": 2.6549907616635964, + "learning_rate": 2.539717736776241e-07, + "loss": 0.9564, + "step": 7008 + }, + { + "epoch": 0.8427824204893886, + "grad_norm": 1.4692997816556745, + "learning_rate": 2.535920055890097e-07, + "loss": 0.9991, + "step": 7009 + }, + { + "epoch": 0.8429026633800276, + "grad_norm": 8.70216986245048, + "learning_rate": 2.5321250242910006e-07, + "loss": 0.8806, + "step": 7010 + }, + { + "epoch": 0.8430229062706668, + "grad_norm": 1.7042450327991803, + "learning_rate": 2.5283326425546493e-07, + "loss": 1.102, + "step": 7011 + }, + { + "epoch": 0.8431431491613058, + "grad_norm": 2.0232689982476426, + "learning_rate": 2.5245429112563443e-07, + "loss": 0.9354, + "step": 7012 + }, + { + "epoch": 0.8432633920519449, + "grad_norm": 1.9170248346565737, + "learning_rate": 2.5207558309709865e-07, + "loss": 1.0583, + "step": 7013 + }, + { + "epoch": 0.8433836349425841, + "grad_norm": 0.6673160509538616, + "learning_rate": 2.516971402273065e-07, + "loss": 0.8, + "step": 7014 + }, + { + "epoch": 0.8435038778332231, + "grad_norm": 2.0301226495231597, + "learning_rate": 2.513189625736687e-07, + "loss": 0.9111, + "step": 7015 + }, + { + "epoch": 0.8436241207238622, + "grad_norm": 2.441138168875288, + "learning_rate": 2.509410501935534e-07, + "loss": 0.9553, + "step": 7016 + }, + { + "epoch": 0.8437443636145013, + "grad_norm": 2.5518030337076087, + "learning_rate": 2.5056340314429116e-07, + "loss": 0.9882, + "step": 7017 + }, + { + "epoch": 0.8438646065051404, + "grad_norm": 2.920487181376921, + "learning_rate": 2.5018602148316904e-07, + "loss": 1.0397, + "step": 7018 + }, + { + "epoch": 0.8439848493957794, + "grad_norm": 1.7739223540042244, + "learning_rate": 2.498089052674359e-07, + "loss": 1.0399, + "step": 7019 + }, + { + "epoch": 0.8441050922864186, + "grad_norm": 1.9921543492569616, + "learning_rate": 2.494320545543007e-07, + "loss": 0.9871, + "step": 7020 + }, + { + "epoch": 0.8442253351770577, + "grad_norm": 2.0199696638643188, + "learning_rate": 2.490554694009308e-07, + "loss": 0.905, + "step": 7021 + }, + { + "epoch": 0.8443455780676967, + "grad_norm": 1.735509827652221, + "learning_rate": 2.4867914986445426e-07, + "loss": 1.0212, + "step": 7022 + }, + { + "epoch": 0.8444658209583359, + "grad_norm": 2.1606275930656693, + "learning_rate": 2.483030960019581e-07, + "loss": 0.9514, + "step": 7023 + }, + { + "epoch": 0.8445860638489749, + "grad_norm": 0.755851331564519, + "learning_rate": 2.479273078704891e-07, + "loss": 0.7928, + "step": 7024 + }, + { + "epoch": 0.844706306739614, + "grad_norm": 0.790387957173372, + "learning_rate": 2.475517855270552e-07, + "loss": 0.8848, + "step": 7025 + }, + { + "epoch": 0.8448265496302532, + "grad_norm": 2.111182900373761, + "learning_rate": 2.4717652902862143e-07, + "loss": 0.9626, + "step": 7026 + }, + { + "epoch": 0.8449467925208922, + "grad_norm": 1.7607392085214677, + "learning_rate": 2.4680153843211495e-07, + "loss": 1.0522, + "step": 7027 + }, + { + "epoch": 0.8450670354115313, + "grad_norm": 2.002539453010666, + "learning_rate": 2.464268137944212e-07, + "loss": 0.9596, + "step": 7028 + }, + { + "epoch": 0.8451872783021703, + "grad_norm": 2.2465213799782684, + "learning_rate": 2.46052355172385e-07, + "loss": 1.0215, + "step": 7029 + }, + { + "epoch": 0.8453075211928095, + "grad_norm": 1.8258451281673953, + "learning_rate": 2.456781626228128e-07, + "loss": 0.9848, + "step": 7030 + }, + { + "epoch": 0.8454277640834486, + "grad_norm": 0.9286700218646236, + "learning_rate": 2.453042362024675e-07, + "loss": 0.8999, + "step": 7031 + }, + { + "epoch": 0.8455480069740876, + "grad_norm": 1.7750866989668073, + "learning_rate": 2.449305759680751e-07, + "loss": 0.9702, + "step": 7032 + }, + { + "epoch": 0.8456682498647268, + "grad_norm": 1.5994080730088045, + "learning_rate": 2.445571819763188e-07, + "loss": 0.9879, + "step": 7033 + }, + { + "epoch": 0.8457884927553658, + "grad_norm": 1.6435659298254461, + "learning_rate": 2.4418405428384227e-07, + "loss": 0.817, + "step": 7034 + }, + { + "epoch": 0.8459087356460049, + "grad_norm": 1.7390946788843968, + "learning_rate": 2.4381119294724864e-07, + "loss": 0.951, + "step": 7035 + }, + { + "epoch": 0.846028978536644, + "grad_norm": 2.472074756677091, + "learning_rate": 2.434385980231004e-07, + "loss": 0.7704, + "step": 7036 + }, + { + "epoch": 0.8461492214272831, + "grad_norm": 1.6301365146925006, + "learning_rate": 2.4306626956792043e-07, + "loss": 0.8928, + "step": 7037 + }, + { + "epoch": 0.8462694643179222, + "grad_norm": 1.8305803422048887, + "learning_rate": 2.4269420763819017e-07, + "loss": 0.9916, + "step": 7038 + }, + { + "epoch": 0.8463897072085613, + "grad_norm": 2.6129471318662807, + "learning_rate": 2.4232241229035223e-07, + "loss": 1.0689, + "step": 7039 + }, + { + "epoch": 0.8465099500992004, + "grad_norm": 0.758579023789885, + "learning_rate": 2.419508835808064e-07, + "loss": 0.8044, + "step": 7040 + }, + { + "epoch": 0.8466301929898394, + "grad_norm": 1.9655190249744776, + "learning_rate": 2.415796215659134e-07, + "loss": 0.8691, + "step": 7041 + }, + { + "epoch": 0.8467504358804786, + "grad_norm": 2.444586210282213, + "learning_rate": 2.412086263019939e-07, + "loss": 1.0077, + "step": 7042 + }, + { + "epoch": 0.8468706787711177, + "grad_norm": 4.120152486182934, + "learning_rate": 2.408378978453276e-07, + "loss": 1.0376, + "step": 7043 + }, + { + "epoch": 0.8469909216617567, + "grad_norm": 0.8244396835644102, + "learning_rate": 2.404674362521533e-07, + "loss": 0.8769, + "step": 7044 + }, + { + "epoch": 0.8471111645523959, + "grad_norm": 2.727406106201596, + "learning_rate": 2.4009724157866997e-07, + "loss": 0.9822, + "step": 7045 + }, + { + "epoch": 0.8472314074430349, + "grad_norm": 2.1226009964025736, + "learning_rate": 2.3972731388103564e-07, + "loss": 1.0052, + "step": 7046 + }, + { + "epoch": 0.847351650333674, + "grad_norm": 0.8159814121639664, + "learning_rate": 2.393576532153687e-07, + "loss": 0.8612, + "step": 7047 + }, + { + "epoch": 0.8474718932243132, + "grad_norm": 0.9358998311762143, + "learning_rate": 2.389882596377453e-07, + "loss": 0.8152, + "step": 7048 + }, + { + "epoch": 0.8475921361149522, + "grad_norm": 1.867688355247706, + "learning_rate": 2.386191332042031e-07, + "loss": 1.0041, + "step": 7049 + }, + { + "epoch": 0.8477123790055913, + "grad_norm": 1.8021083087697214, + "learning_rate": 2.3825027397073794e-07, + "loss": 0.9651, + "step": 7050 + }, + { + "epoch": 0.8478326218962304, + "grad_norm": 2.0465494144456473, + "learning_rate": 2.3788168199330515e-07, + "loss": 0.8989, + "step": 7051 + }, + { + "epoch": 0.8479528647868695, + "grad_norm": 1.7749493578281648, + "learning_rate": 2.3751335732782074e-07, + "loss": 0.9639, + "step": 7052 + }, + { + "epoch": 0.8480731076775085, + "grad_norm": 1.9959823069657854, + "learning_rate": 2.371453000301582e-07, + "loss": 1.0344, + "step": 7053 + }, + { + "epoch": 0.8481933505681477, + "grad_norm": 2.1901857509831935, + "learning_rate": 2.3677751015615222e-07, + "loss": 0.9803, + "step": 7054 + }, + { + "epoch": 0.8483135934587868, + "grad_norm": 1.786687749102503, + "learning_rate": 2.3640998776159593e-07, + "loss": 1.0898, + "step": 7055 + }, + { + "epoch": 0.8484338363494258, + "grad_norm": 1.702501235266866, + "learning_rate": 2.3604273290224253e-07, + "loss": 1.0476, + "step": 7056 + }, + { + "epoch": 0.848554079240065, + "grad_norm": 2.720870320261996, + "learning_rate": 2.356757456338039e-07, + "loss": 0.9834, + "step": 7057 + }, + { + "epoch": 0.848674322130704, + "grad_norm": 0.7540751450858115, + "learning_rate": 2.3530902601195147e-07, + "loss": 0.8268, + "step": 7058 + }, + { + "epoch": 0.8487945650213431, + "grad_norm": 2.8900687017491604, + "learning_rate": 2.34942574092317e-07, + "loss": 1.0276, + "step": 7059 + }, + { + "epoch": 0.8489148079119821, + "grad_norm": 2.1623526197717817, + "learning_rate": 2.3457638993049045e-07, + "loss": 1.0035, + "step": 7060 + }, + { + "epoch": 0.8490350508026213, + "grad_norm": 1.874882694389587, + "learning_rate": 2.3421047358202252e-07, + "loss": 0.8762, + "step": 7061 + }, + { + "epoch": 0.8491552936932604, + "grad_norm": 2.5411654640018604, + "learning_rate": 2.3384482510242144e-07, + "loss": 1.0648, + "step": 7062 + }, + { + "epoch": 0.8492755365838994, + "grad_norm": 1.8710143643862016, + "learning_rate": 2.3347944454715575e-07, + "loss": 1.0111, + "step": 7063 + }, + { + "epoch": 0.8493957794745386, + "grad_norm": 3.6080334304563375, + "learning_rate": 2.331143319716542e-07, + "loss": 0.9114, + "step": 7064 + }, + { + "epoch": 0.8495160223651776, + "grad_norm": 2.490354910532597, + "learning_rate": 2.3274948743130363e-07, + "loss": 0.8954, + "step": 7065 + }, + { + "epoch": 0.8496362652558167, + "grad_norm": 1.7264164651414358, + "learning_rate": 2.3238491098145085e-07, + "loss": 1.0307, + "step": 7066 + }, + { + "epoch": 0.8497565081464559, + "grad_norm": 2.237120748590566, + "learning_rate": 2.3202060267740141e-07, + "loss": 0.9705, + "step": 7067 + }, + { + "epoch": 0.8498767510370949, + "grad_norm": 3.5610261511888233, + "learning_rate": 2.3165656257442044e-07, + "loss": 1.0058, + "step": 7068 + }, + { + "epoch": 0.849996993927734, + "grad_norm": 1.8972838864499089, + "learning_rate": 2.31292790727734e-07, + "loss": 1.139, + "step": 7069 + }, + { + "epoch": 0.8501172368183731, + "grad_norm": 2.221567755593804, + "learning_rate": 2.3092928719252392e-07, + "loss": 1.0401, + "step": 7070 + }, + { + "epoch": 0.8502374797090122, + "grad_norm": 2.098436263844573, + "learning_rate": 2.3056605202393475e-07, + "loss": 1.0182, + "step": 7071 + }, + { + "epoch": 0.8503577225996513, + "grad_norm": 1.8855939398898407, + "learning_rate": 2.3020308527706888e-07, + "loss": 0.9046, + "step": 7072 + }, + { + "epoch": 0.8504779654902904, + "grad_norm": 1.855942579301961, + "learning_rate": 2.2984038700698715e-07, + "loss": 1.1235, + "step": 7073 + }, + { + "epoch": 0.8505982083809295, + "grad_norm": 1.755687191745383, + "learning_rate": 2.2947795726871222e-07, + "loss": 1.0276, + "step": 7074 + }, + { + "epoch": 0.8507184512715685, + "grad_norm": 2.058616913116839, + "learning_rate": 2.2911579611722253e-07, + "loss": 1.095, + "step": 7075 + }, + { + "epoch": 0.8508386941622077, + "grad_norm": 1.824201105918754, + "learning_rate": 2.2875390360745905e-07, + "loss": 1.1098, + "step": 7076 + }, + { + "epoch": 0.8509589370528468, + "grad_norm": 1.7037257963435861, + "learning_rate": 2.2839227979432008e-07, + "loss": 1.0103, + "step": 7077 + }, + { + "epoch": 0.8510791799434858, + "grad_norm": 2.0190185840614854, + "learning_rate": 2.2803092473266373e-07, + "loss": 1.0886, + "step": 7078 + }, + { + "epoch": 0.851199422834125, + "grad_norm": 2.3314359534504705, + "learning_rate": 2.2766983847730724e-07, + "loss": 1.1002, + "step": 7079 + }, + { + "epoch": 0.851319665724764, + "grad_norm": 2.1368472881813085, + "learning_rate": 2.2730902108302663e-07, + "loss": 0.9028, + "step": 7080 + }, + { + "epoch": 0.8514399086154031, + "grad_norm": 1.6753398607407444, + "learning_rate": 2.269484726045583e-07, + "loss": 0.9244, + "step": 7081 + }, + { + "epoch": 0.8515601515060423, + "grad_norm": 3.009997399161093, + "learning_rate": 2.2658819309659672e-07, + "loss": 1.0262, + "step": 7082 + }, + { + "epoch": 0.8516803943966813, + "grad_norm": 1.9894049865782881, + "learning_rate": 2.2622818261379706e-07, + "loss": 1.0864, + "step": 7083 + }, + { + "epoch": 0.8518006372873204, + "grad_norm": 4.440168411222657, + "learning_rate": 2.2586844121077142e-07, + "loss": 0.986, + "step": 7084 + }, + { + "epoch": 0.8519208801779595, + "grad_norm": 2.0502798866314302, + "learning_rate": 2.2550896894209215e-07, + "loss": 0.9553, + "step": 7085 + }, + { + "epoch": 0.8520411230685986, + "grad_norm": 0.690900822618795, + "learning_rate": 2.2514976586229184e-07, + "loss": 0.8032, + "step": 7086 + }, + { + "epoch": 0.8521613659592376, + "grad_norm": 0.7778939068663626, + "learning_rate": 2.247908320258609e-07, + "loss": 0.7842, + "step": 7087 + }, + { + "epoch": 0.8522816088498768, + "grad_norm": 2.3854245736825375, + "learning_rate": 2.2443216748724914e-07, + "loss": 1.0349, + "step": 7088 + }, + { + "epoch": 0.8524018517405159, + "grad_norm": 1.950225753431988, + "learning_rate": 2.2407377230086588e-07, + "loss": 0.9808, + "step": 7089 + }, + { + "epoch": 0.8525220946311549, + "grad_norm": 2.5117083085052214, + "learning_rate": 2.23715646521079e-07, + "loss": 1.0688, + "step": 7090 + }, + { + "epoch": 0.852642337521794, + "grad_norm": 2.0916492684338577, + "learning_rate": 2.2335779020221724e-07, + "loss": 1.0791, + "step": 7091 + }, + { + "epoch": 0.8527625804124331, + "grad_norm": 0.8108737148711651, + "learning_rate": 2.2300020339856497e-07, + "loss": 0.8023, + "step": 7092 + }, + { + "epoch": 0.8528828233030722, + "grad_norm": 2.193272630533684, + "learning_rate": 2.2264288616436966e-07, + "loss": 1.0122, + "step": 7093 + }, + { + "epoch": 0.8530030661937112, + "grad_norm": 2.0162086528930314, + "learning_rate": 2.222858385538351e-07, + "loss": 0.9636, + "step": 7094 + }, + { + "epoch": 0.8531233090843504, + "grad_norm": 1.935319596543522, + "learning_rate": 2.2192906062112527e-07, + "loss": 0.9186, + "step": 7095 + }, + { + "epoch": 0.8532435519749895, + "grad_norm": 2.0803926185375743, + "learning_rate": 2.2157255242036377e-07, + "loss": 0.9408, + "step": 7096 + }, + { + "epoch": 0.8533637948656285, + "grad_norm": 2.208209400178208, + "learning_rate": 2.2121631400563135e-07, + "loss": 0.9827, + "step": 7097 + }, + { + "epoch": 0.8534840377562677, + "grad_norm": 0.777775017037109, + "learning_rate": 2.208603454309701e-07, + "loss": 0.7666, + "step": 7098 + }, + { + "epoch": 0.8536042806469067, + "grad_norm": 2.0151675392716677, + "learning_rate": 2.2050464675037994e-07, + "loss": 0.9451, + "step": 7099 + }, + { + "epoch": 0.8537245235375458, + "grad_norm": 2.7139193710685534, + "learning_rate": 2.2014921801782016e-07, + "loss": 0.9656, + "step": 7100 + }, + { + "epoch": 0.853844766428185, + "grad_norm": 1.92998635143929, + "learning_rate": 2.1979405928720872e-07, + "loss": 0.9759, + "step": 7101 + }, + { + "epoch": 0.853965009318824, + "grad_norm": 1.4797817437875176, + "learning_rate": 2.1943917061242257e-07, + "loss": 1.0326, + "step": 7102 + }, + { + "epoch": 0.8540852522094631, + "grad_norm": 1.6817603517238993, + "learning_rate": 2.1908455204729903e-07, + "loss": 0.899, + "step": 7103 + }, + { + "epoch": 0.8542054951001022, + "grad_norm": 2.275930604181786, + "learning_rate": 2.1873020364563265e-07, + "loss": 1.0193, + "step": 7104 + }, + { + "epoch": 0.8543257379907413, + "grad_norm": 2.3207974424845803, + "learning_rate": 2.183761254611789e-07, + "loss": 0.9961, + "step": 7105 + }, + { + "epoch": 0.8544459808813804, + "grad_norm": 1.8348721274119477, + "learning_rate": 2.1802231754764987e-07, + "loss": 0.9372, + "step": 7106 + }, + { + "epoch": 0.8545662237720195, + "grad_norm": 1.845184091724098, + "learning_rate": 2.17668779958718e-07, + "loss": 0.9969, + "step": 7107 + }, + { + "epoch": 0.8546864666626586, + "grad_norm": 2.7558581638130755, + "learning_rate": 2.1731551274801553e-07, + "loss": 1.0367, + "step": 7108 + }, + { + "epoch": 0.8548067095532976, + "grad_norm": 3.3128015632833896, + "learning_rate": 2.169625159691324e-07, + "loss": 0.8493, + "step": 7109 + }, + { + "epoch": 0.8549269524439368, + "grad_norm": 4.867340526348909, + "learning_rate": 2.1660978967561784e-07, + "loss": 0.982, + "step": 7110 + }, + { + "epoch": 0.8550471953345758, + "grad_norm": 2.46799988758699, + "learning_rate": 2.1625733392098035e-07, + "loss": 1.0278, + "step": 7111 + }, + { + "epoch": 0.8551674382252149, + "grad_norm": 1.8948597416769537, + "learning_rate": 2.159051487586867e-07, + "loss": 1.0309, + "step": 7112 + }, + { + "epoch": 0.8552876811158541, + "grad_norm": 3.5241048988551356, + "learning_rate": 2.155532342421642e-07, + "loss": 0.9576, + "step": 7113 + }, + { + "epoch": 0.8554079240064931, + "grad_norm": 3.012335846413363, + "learning_rate": 2.1520159042479636e-07, + "loss": 1.0208, + "step": 7114 + }, + { + "epoch": 0.8555281668971322, + "grad_norm": 4.690998043751923, + "learning_rate": 2.148502173599287e-07, + "loss": 0.9469, + "step": 7115 + }, + { + "epoch": 0.8556484097877713, + "grad_norm": 3.2577158961621286, + "learning_rate": 2.1449911510086372e-07, + "loss": 0.8986, + "step": 7116 + }, + { + "epoch": 0.8557686526784104, + "grad_norm": 1.9478113754565431, + "learning_rate": 2.141482837008628e-07, + "loss": 1.0043, + "step": 7117 + }, + { + "epoch": 0.8558888955690495, + "grad_norm": 1.8648127693508232, + "learning_rate": 2.1379772321314826e-07, + "loss": 0.9585, + "step": 7118 + }, + { + "epoch": 0.8560091384596886, + "grad_norm": 1.9947006828999614, + "learning_rate": 2.1344743369089802e-07, + "loss": 1.0575, + "step": 7119 + }, + { + "epoch": 0.8561293813503277, + "grad_norm": 1.7636914764362714, + "learning_rate": 2.130974151872522e-07, + "loss": 1.0594, + "step": 7120 + }, + { + "epoch": 0.8562496242409667, + "grad_norm": 1.6946345085456365, + "learning_rate": 2.1274766775530773e-07, + "loss": 1.0231, + "step": 7121 + }, + { + "epoch": 0.8563698671316058, + "grad_norm": 2.1789822001331984, + "learning_rate": 2.1239819144812077e-07, + "loss": 1.0332, + "step": 7122 + }, + { + "epoch": 0.856490110022245, + "grad_norm": 2.082881675080871, + "learning_rate": 2.1204898631870716e-07, + "loss": 0.9381, + "step": 7123 + }, + { + "epoch": 0.856610352912884, + "grad_norm": 2.0043194142250638, + "learning_rate": 2.1170005242004006e-07, + "loss": 0.9993, + "step": 7124 + }, + { + "epoch": 0.8567305958035231, + "grad_norm": 1.9540300027306785, + "learning_rate": 2.1135138980505384e-07, + "loss": 1.0164, + "step": 7125 + }, + { + "epoch": 0.8568508386941622, + "grad_norm": 1.7387924299952437, + "learning_rate": 2.110029985266395e-07, + "loss": 0.9623, + "step": 7126 + }, + { + "epoch": 0.8569710815848013, + "grad_norm": 2.0790211745776044, + "learning_rate": 2.1065487863764787e-07, + "loss": 0.974, + "step": 7127 + }, + { + "epoch": 0.8570913244754403, + "grad_norm": 1.6869373089821789, + "learning_rate": 2.1030703019088846e-07, + "loss": 1.0937, + "step": 7128 + }, + { + "epoch": 0.8572115673660795, + "grad_norm": 1.8831821569459561, + "learning_rate": 2.099594532391291e-07, + "loss": 0.9427, + "step": 7129 + }, + { + "epoch": 0.8573318102567186, + "grad_norm": 1.6269008083106318, + "learning_rate": 2.0961214783509806e-07, + "loss": 1.0244, + "step": 7130 + }, + { + "epoch": 0.8574520531473576, + "grad_norm": 1.8457896701766325, + "learning_rate": 2.0926511403148051e-07, + "loss": 0.9857, + "step": 7131 + }, + { + "epoch": 0.8575722960379968, + "grad_norm": 2.112296786494123, + "learning_rate": 2.0891835188092143e-07, + "loss": 0.9975, + "step": 7132 + }, + { + "epoch": 0.8576925389286358, + "grad_norm": 1.8286209737822177, + "learning_rate": 2.0857186143602434e-07, + "loss": 1.055, + "step": 7133 + }, + { + "epoch": 0.8578127818192749, + "grad_norm": 1.7826501797494076, + "learning_rate": 2.0822564274935094e-07, + "loss": 0.9099, + "step": 7134 + }, + { + "epoch": 0.8579330247099141, + "grad_norm": 1.9473682862426906, + "learning_rate": 2.078796958734239e-07, + "loss": 0.9039, + "step": 7135 + }, + { + "epoch": 0.8580532676005531, + "grad_norm": 2.033394136718329, + "learning_rate": 2.0753402086072124e-07, + "loss": 0.9859, + "step": 7136 + }, + { + "epoch": 0.8581735104911922, + "grad_norm": 3.8977446465973276, + "learning_rate": 2.071886177636828e-07, + "loss": 0.989, + "step": 7137 + }, + { + "epoch": 0.8582937533818313, + "grad_norm": 1.8235514888748041, + "learning_rate": 2.0684348663470575e-07, + "loss": 1.0652, + "step": 7138 + }, + { + "epoch": 0.8584139962724704, + "grad_norm": 2.5134197544380665, + "learning_rate": 2.0649862752614555e-07, + "loss": 0.8506, + "step": 7139 + }, + { + "epoch": 0.8585342391631094, + "grad_norm": 0.7667119620943732, + "learning_rate": 2.0615404049031838e-07, + "loss": 0.8079, + "step": 7140 + }, + { + "epoch": 0.8586544820537486, + "grad_norm": 2.535144776493201, + "learning_rate": 2.0580972557949616e-07, + "loss": 1.016, + "step": 7141 + }, + { + "epoch": 0.8587747249443877, + "grad_norm": 0.8087615792626215, + "learning_rate": 2.054656828459125e-07, + "loss": 0.7796, + "step": 7142 + }, + { + "epoch": 0.8588949678350267, + "grad_norm": 1.8233961582809066, + "learning_rate": 2.051219123417578e-07, + "loss": 1.0104, + "step": 7143 + }, + { + "epoch": 0.8590152107256659, + "grad_norm": 2.214555563502879, + "learning_rate": 2.0477841411918196e-07, + "loss": 0.8361, + "step": 7144 + }, + { + "epoch": 0.859135453616305, + "grad_norm": 11.23595857667332, + "learning_rate": 2.0443518823029326e-07, + "loss": 0.9795, + "step": 7145 + }, + { + "epoch": 0.859255696506944, + "grad_norm": 2.0472784445553605, + "learning_rate": 2.0409223472715854e-07, + "loss": 1.0042, + "step": 7146 + }, + { + "epoch": 0.8593759393975832, + "grad_norm": 2.13595483712874, + "learning_rate": 2.0374955366180434e-07, + "loss": 0.989, + "step": 7147 + }, + { + "epoch": 0.8594961822882222, + "grad_norm": 1.78236347665452, + "learning_rate": 2.034071450862147e-07, + "loss": 0.9652, + "step": 7148 + }, + { + "epoch": 0.8596164251788613, + "grad_norm": 1.9584304771948728, + "learning_rate": 2.030650090523327e-07, + "loss": 1.0073, + "step": 7149 + }, + { + "epoch": 0.8597366680695004, + "grad_norm": 1.7450099372034307, + "learning_rate": 2.0272314561205995e-07, + "loss": 0.8339, + "step": 7150 + }, + { + "epoch": 0.8598569109601395, + "grad_norm": 1.8669598811762809, + "learning_rate": 2.023815548172567e-07, + "loss": 0.9675, + "step": 7151 + }, + { + "epoch": 0.8599771538507786, + "grad_norm": 1.6199497828529914, + "learning_rate": 2.0204023671974267e-07, + "loss": 0.8962, + "step": 7152 + }, + { + "epoch": 0.8600973967414177, + "grad_norm": 2.151124030530681, + "learning_rate": 2.0169919137129532e-07, + "loss": 1.047, + "step": 7153 + }, + { + "epoch": 0.8602176396320568, + "grad_norm": 3.6769538979087284, + "learning_rate": 2.013584188236508e-07, + "loss": 0.9428, + "step": 7154 + }, + { + "epoch": 0.8603378825226958, + "grad_norm": 2.0072652464313854, + "learning_rate": 2.0101791912850396e-07, + "loss": 1.0301, + "step": 7155 + }, + { + "epoch": 0.8604581254133349, + "grad_norm": 1.8769439792310998, + "learning_rate": 2.006776923375082e-07, + "loss": 0.8738, + "step": 7156 + }, + { + "epoch": 0.860578368303974, + "grad_norm": 1.5966816286781014, + "learning_rate": 2.003377385022764e-07, + "loss": 0.9519, + "step": 7157 + }, + { + "epoch": 0.8606986111946131, + "grad_norm": 2.0659649960118136, + "learning_rate": 1.9999805767437826e-07, + "loss": 1.0119, + "step": 7158 + }, + { + "epoch": 0.8608188540852522, + "grad_norm": 2.1577430752601217, + "learning_rate": 1.9965864990534386e-07, + "loss": 0.9543, + "step": 7159 + }, + { + "epoch": 0.8609390969758913, + "grad_norm": 2.167408693847792, + "learning_rate": 1.9931951524666092e-07, + "loss": 1.0117, + "step": 7160 + }, + { + "epoch": 0.8610593398665304, + "grad_norm": 3.0164961273153326, + "learning_rate": 1.9898065374977534e-07, + "loss": 1.0479, + "step": 7161 + }, + { + "epoch": 0.8611795827571694, + "grad_norm": 2.135301527935541, + "learning_rate": 1.9864206546609342e-07, + "loss": 0.9698, + "step": 7162 + }, + { + "epoch": 0.8612998256478086, + "grad_norm": 1.8842800008889014, + "learning_rate": 1.983037504469771e-07, + "loss": 1.0791, + "step": 7163 + }, + { + "epoch": 0.8614200685384477, + "grad_norm": 1.9795770387008154, + "learning_rate": 1.9796570874374984e-07, + "loss": 0.9048, + "step": 7164 + }, + { + "epoch": 0.8615403114290867, + "grad_norm": 1.8820499126289687, + "learning_rate": 1.976279404076917e-07, + "loss": 1.0139, + "step": 7165 + }, + { + "epoch": 0.8616605543197259, + "grad_norm": 1.7388419972043048, + "learning_rate": 1.9729044549004193e-07, + "loss": 0.9999, + "step": 7166 + }, + { + "epoch": 0.8617807972103649, + "grad_norm": 1.6475358277797023, + "learning_rate": 1.9695322404199822e-07, + "loss": 0.9416, + "step": 7167 + }, + { + "epoch": 0.861901040101004, + "grad_norm": 1.890846806905911, + "learning_rate": 1.9661627611471654e-07, + "loss": 1.0623, + "step": 7168 + }, + { + "epoch": 0.8620212829916432, + "grad_norm": 2.0748765839791816, + "learning_rate": 1.9627960175931246e-07, + "loss": 0.9406, + "step": 7169 + }, + { + "epoch": 0.8621415258822822, + "grad_norm": 1.8492426120909076, + "learning_rate": 1.9594320102685847e-07, + "loss": 0.9759, + "step": 7170 + }, + { + "epoch": 0.8622617687729213, + "grad_norm": 1.88192284803523, + "learning_rate": 1.956070739683864e-07, + "loss": 0.873, + "step": 7171 + }, + { + "epoch": 0.8623820116635604, + "grad_norm": 1.5783802996814054, + "learning_rate": 1.9527122063488678e-07, + "loss": 0.9788, + "step": 7172 + }, + { + "epoch": 0.8625022545541995, + "grad_norm": 1.847437775096965, + "learning_rate": 1.9493564107730755e-07, + "loss": 1.0411, + "step": 7173 + }, + { + "epoch": 0.8626224974448385, + "grad_norm": 1.8740281447390725, + "learning_rate": 1.9460033534655684e-07, + "loss": 0.8501, + "step": 7174 + }, + { + "epoch": 0.8627427403354777, + "grad_norm": 1.549833993328192, + "learning_rate": 1.9426530349349978e-07, + "loss": 1.0774, + "step": 7175 + }, + { + "epoch": 0.8628629832261168, + "grad_norm": 2.0056791735328847, + "learning_rate": 1.9393054556896038e-07, + "loss": 0.8828, + "step": 7176 + }, + { + "epoch": 0.8629832261167558, + "grad_norm": 2.202207225344744, + "learning_rate": 1.9359606162372133e-07, + "loss": 0.9274, + "step": 7177 + }, + { + "epoch": 0.863103469007395, + "grad_norm": 1.8836594572121121, + "learning_rate": 1.9326185170852293e-07, + "loss": 0.9414, + "step": 7178 + }, + { + "epoch": 0.863223711898034, + "grad_norm": 16.193198125365594, + "learning_rate": 1.9292791587406598e-07, + "loss": 0.9603, + "step": 7179 + }, + { + "epoch": 0.8633439547886731, + "grad_norm": 1.9935536207149713, + "learning_rate": 1.9259425417100661e-07, + "loss": 1.104, + "step": 7180 + }, + { + "epoch": 0.8634641976793123, + "grad_norm": 2.3952010495210945, + "learning_rate": 1.9226086664996234e-07, + "loss": 0.9885, + "step": 7181 + }, + { + "epoch": 0.8635844405699513, + "grad_norm": 2.252545188763834, + "learning_rate": 1.9192775336150712e-07, + "loss": 0.9792, + "step": 7182 + }, + { + "epoch": 0.8637046834605904, + "grad_norm": 0.792281054628735, + "learning_rate": 1.915949143561739e-07, + "loss": 0.8006, + "step": 7183 + }, + { + "epoch": 0.8638249263512295, + "grad_norm": 1.8035250796738855, + "learning_rate": 1.9126234968445498e-07, + "loss": 1.0136, + "step": 7184 + }, + { + "epoch": 0.8639451692418686, + "grad_norm": 1.473482744932029, + "learning_rate": 1.9093005939679884e-07, + "loss": 0.9128, + "step": 7185 + }, + { + "epoch": 0.8640654121325076, + "grad_norm": 1.8706906069521723, + "learning_rate": 1.9059804354361452e-07, + "loss": 0.999, + "step": 7186 + }, + { + "epoch": 0.8641856550231467, + "grad_norm": 1.7375703536685223, + "learning_rate": 1.902663021752684e-07, + "loss": 0.9404, + "step": 7187 + }, + { + "epoch": 0.8643058979137859, + "grad_norm": 2.4289957024515454, + "learning_rate": 1.8993483534208556e-07, + "loss": 1.0551, + "step": 7188 + }, + { + "epoch": 0.8644261408044249, + "grad_norm": 2.5206084558316197, + "learning_rate": 1.8960364309434884e-07, + "loss": 0.9886, + "step": 7189 + }, + { + "epoch": 0.864546383695064, + "grad_norm": 1.7435414480102174, + "learning_rate": 1.8927272548229967e-07, + "loss": 1.018, + "step": 7190 + }, + { + "epoch": 0.8646666265857031, + "grad_norm": 1.6153694290104033, + "learning_rate": 1.8894208255613876e-07, + "loss": 1.0707, + "step": 7191 + }, + { + "epoch": 0.8647868694763422, + "grad_norm": 2.125717602905199, + "learning_rate": 1.8861171436602397e-07, + "loss": 1.0155, + "step": 7192 + }, + { + "epoch": 0.8649071123669813, + "grad_norm": 2.342926084022471, + "learning_rate": 1.882816209620719e-07, + "loss": 1.0416, + "step": 7193 + }, + { + "epoch": 0.8650273552576204, + "grad_norm": 1.8054371387416488, + "learning_rate": 1.8795180239435738e-07, + "loss": 1.0067, + "step": 7194 + }, + { + "epoch": 0.8651475981482595, + "grad_norm": 3.335867887748878, + "learning_rate": 1.8762225871291348e-07, + "loss": 1.0038, + "step": 7195 + }, + { + "epoch": 0.8652678410388985, + "grad_norm": 1.728317680447411, + "learning_rate": 1.8729298996773201e-07, + "loss": 1.0467, + "step": 7196 + }, + { + "epoch": 0.8653880839295377, + "grad_norm": 0.847978381286212, + "learning_rate": 1.8696399620876301e-07, + "loss": 0.8481, + "step": 7197 + }, + { + "epoch": 0.8655083268201768, + "grad_norm": 2.1408838268763355, + "learning_rate": 1.866352774859141e-07, + "loss": 1.0305, + "step": 7198 + }, + { + "epoch": 0.8656285697108158, + "grad_norm": 2.074672038806017, + "learning_rate": 1.8630683384905188e-07, + "loss": 0.9263, + "step": 7199 + }, + { + "epoch": 0.865748812601455, + "grad_norm": 1.9312685153900757, + "learning_rate": 1.8597866534800045e-07, + "loss": 1.1216, + "step": 7200 + }, + { + "epoch": 0.865869055492094, + "grad_norm": 2.138355137611515, + "learning_rate": 1.8565077203254398e-07, + "loss": 0.982, + "step": 7201 + }, + { + "epoch": 0.8659892983827331, + "grad_norm": 7.748488740315047, + "learning_rate": 1.8532315395242203e-07, + "loss": 0.9606, + "step": 7202 + }, + { + "epoch": 0.8661095412733723, + "grad_norm": 2.0815983477891584, + "learning_rate": 1.849958111573353e-07, + "loss": 0.9597, + "step": 7203 + }, + { + "epoch": 0.8662297841640113, + "grad_norm": 1.77492726136653, + "learning_rate": 1.8466874369694074e-07, + "loss": 0.8775, + "step": 7204 + }, + { + "epoch": 0.8663500270546504, + "grad_norm": 2.427631845557204, + "learning_rate": 1.843419516208542e-07, + "loss": 0.9387, + "step": 7205 + }, + { + "epoch": 0.8664702699452895, + "grad_norm": 2.98909494130167, + "learning_rate": 1.8401543497865047e-07, + "loss": 1.0339, + "step": 7206 + }, + { + "epoch": 0.8665905128359286, + "grad_norm": 2.6385356153520925, + "learning_rate": 1.836891938198608e-07, + "loss": 0.8784, + "step": 7207 + }, + { + "epoch": 0.8667107557265676, + "grad_norm": 2.2674346010982473, + "learning_rate": 1.8336322819397677e-07, + "loss": 0.9505, + "step": 7208 + }, + { + "epoch": 0.8668309986172068, + "grad_norm": 1.8713533543932588, + "learning_rate": 1.8303753815044654e-07, + "loss": 0.8607, + "step": 7209 + }, + { + "epoch": 0.8669512415078459, + "grad_norm": 2.506931138172101, + "learning_rate": 1.827121237386773e-07, + "loss": 0.94, + "step": 7210 + }, + { + "epoch": 0.8670714843984849, + "grad_norm": 2.53692537011973, + "learning_rate": 1.8238698500803374e-07, + "loss": 0.9873, + "step": 7211 + }, + { + "epoch": 0.8671917272891241, + "grad_norm": 0.7418714954836567, + "learning_rate": 1.820621220078391e-07, + "loss": 0.8, + "step": 7212 + }, + { + "epoch": 0.8673119701797631, + "grad_norm": 1.859514311372028, + "learning_rate": 1.8173753478737553e-07, + "loss": 0.9161, + "step": 7213 + }, + { + "epoch": 0.8674322130704022, + "grad_norm": 2.2006848396414544, + "learning_rate": 1.8141322339588205e-07, + "loss": 1.0381, + "step": 7214 + }, + { + "epoch": 0.8675524559610414, + "grad_norm": 1.999781358211906, + "learning_rate": 1.810891878825569e-07, + "loss": 0.9374, + "step": 7215 + }, + { + "epoch": 0.8676726988516804, + "grad_norm": 2.0281702021894197, + "learning_rate": 1.8076542829655561e-07, + "loss": 0.9543, + "step": 7216 + }, + { + "epoch": 0.8677929417423195, + "grad_norm": 2.1850063824942314, + "learning_rate": 1.8044194468699203e-07, + "loss": 1.03, + "step": 7217 + }, + { + "epoch": 0.8679131846329585, + "grad_norm": 10.141887004363504, + "learning_rate": 1.8011873710293912e-07, + "loss": 0.9938, + "step": 7218 + }, + { + "epoch": 0.8680334275235977, + "grad_norm": 10.709393835927775, + "learning_rate": 1.7979580559342677e-07, + "loss": 0.9309, + "step": 7219 + }, + { + "epoch": 0.8681536704142367, + "grad_norm": 1.7727653809432797, + "learning_rate": 1.7947315020744358e-07, + "loss": 0.9027, + "step": 7220 + }, + { + "epoch": 0.8682739133048758, + "grad_norm": 1.888280881314287, + "learning_rate": 1.7915077099393594e-07, + "loss": 1.041, + "step": 7221 + }, + { + "epoch": 0.868394156195515, + "grad_norm": 1.8278431471343823, + "learning_rate": 1.788286680018083e-07, + "loss": 0.9685, + "step": 7222 + }, + { + "epoch": 0.868514399086154, + "grad_norm": 1.6268043037849114, + "learning_rate": 1.7850684127992443e-07, + "loss": 0.9584, + "step": 7223 + }, + { + "epoch": 0.8686346419767931, + "grad_norm": 1.6423688067990048, + "learning_rate": 1.7818529087710378e-07, + "loss": 0.9421, + "step": 7224 + }, + { + "epoch": 0.8687548848674322, + "grad_norm": 1.8570538305595634, + "learning_rate": 1.7786401684212637e-07, + "loss": 1.0802, + "step": 7225 + }, + { + "epoch": 0.8688751277580713, + "grad_norm": 0.7542099810292631, + "learning_rate": 1.7754301922372883e-07, + "loss": 0.7968, + "step": 7226 + }, + { + "epoch": 0.8689953706487104, + "grad_norm": 1.8710608748253612, + "learning_rate": 1.7722229807060617e-07, + "loss": 1.0465, + "step": 7227 + }, + { + "epoch": 0.8691156135393495, + "grad_norm": 2.21713757018525, + "learning_rate": 1.7690185343141172e-07, + "loss": 1.0556, + "step": 7228 + }, + { + "epoch": 0.8692358564299886, + "grad_norm": 1.9854436685087626, + "learning_rate": 1.7658168535475615e-07, + "loss": 0.9348, + "step": 7229 + }, + { + "epoch": 0.8693560993206276, + "grad_norm": 2.1283027563485097, + "learning_rate": 1.7626179388920948e-07, + "loss": 0.8824, + "step": 7230 + }, + { + "epoch": 0.8694763422112668, + "grad_norm": 1.6556432800320642, + "learning_rate": 1.7594217908329866e-07, + "loss": 1.0431, + "step": 7231 + }, + { + "epoch": 0.8695965851019059, + "grad_norm": 1.754635461730271, + "learning_rate": 1.7562284098550895e-07, + "loss": 0.9753, + "step": 7232 + }, + { + "epoch": 0.8697168279925449, + "grad_norm": 0.8452012332589521, + "learning_rate": 1.753037796442838e-07, + "loss": 0.8597, + "step": 7233 + }, + { + "epoch": 0.8698370708831841, + "grad_norm": 2.2326082300349026, + "learning_rate": 1.74984995108024e-07, + "loss": 0.9931, + "step": 7234 + }, + { + "epoch": 0.8699573137738231, + "grad_norm": 1.949501338190254, + "learning_rate": 1.7466648742508981e-07, + "loss": 1.0667, + "step": 7235 + }, + { + "epoch": 0.8700775566644622, + "grad_norm": 2.193095869912874, + "learning_rate": 1.7434825664379837e-07, + "loss": 1.0827, + "step": 7236 + }, + { + "epoch": 0.8701977995551013, + "grad_norm": 2.6569023677589243, + "learning_rate": 1.740303028124246e-07, + "loss": 1.0962, + "step": 7237 + }, + { + "epoch": 0.8703180424457404, + "grad_norm": 2.4374346602069807, + "learning_rate": 1.7371262597920212e-07, + "loss": 0.9988, + "step": 7238 + }, + { + "epoch": 0.8704382853363795, + "grad_norm": 1.5023297654888164, + "learning_rate": 1.7339522619232195e-07, + "loss": 1.0006, + "step": 7239 + }, + { + "epoch": 0.8705585282270186, + "grad_norm": 2.1626538260549464, + "learning_rate": 1.730781034999338e-07, + "loss": 0.9907, + "step": 7240 + }, + { + "epoch": 0.8706787711176577, + "grad_norm": 2.4911630850950983, + "learning_rate": 1.7276125795014497e-07, + "loss": 0.9729, + "step": 7241 + }, + { + "epoch": 0.8707990140082967, + "grad_norm": 1.875775494739453, + "learning_rate": 1.7244468959102054e-07, + "loss": 0.912, + "step": 7242 + }, + { + "epoch": 0.8709192568989359, + "grad_norm": 2.249276680712105, + "learning_rate": 1.7212839847058348e-07, + "loss": 1.0873, + "step": 7243 + }, + { + "epoch": 0.871039499789575, + "grad_norm": 2.161351457887337, + "learning_rate": 1.718123846368147e-07, + "loss": 0.9741, + "step": 7244 + }, + { + "epoch": 0.871159742680214, + "grad_norm": 1.7158014842494633, + "learning_rate": 1.714966481376543e-07, + "loss": 0.9525, + "step": 7245 + }, + { + "epoch": 0.8712799855708532, + "grad_norm": 1.8933539753894535, + "learning_rate": 1.7118118902099797e-07, + "loss": 1.0694, + "step": 7246 + }, + { + "epoch": 0.8714002284614922, + "grad_norm": 1.7560887680166326, + "learning_rate": 1.7086600733470146e-07, + "loss": 1.0444, + "step": 7247 + }, + { + "epoch": 0.8715204713521313, + "grad_norm": 1.8493663785320305, + "learning_rate": 1.7055110312657738e-07, + "loss": 1.0061, + "step": 7248 + }, + { + "epoch": 0.8716407142427703, + "grad_norm": 2.964525288603148, + "learning_rate": 1.702364764443962e-07, + "loss": 0.9809, + "step": 7249 + }, + { + "epoch": 0.8717609571334095, + "grad_norm": 1.94908266987654, + "learning_rate": 1.6992212733588685e-07, + "loss": 0.9599, + "step": 7250 + }, + { + "epoch": 0.8718812000240486, + "grad_norm": 1.890077306690259, + "learning_rate": 1.6960805584873538e-07, + "loss": 0.9844, + "step": 7251 + }, + { + "epoch": 0.8720014429146876, + "grad_norm": 1.7981861140977142, + "learning_rate": 1.6929426203058684e-07, + "loss": 1.0195, + "step": 7252 + }, + { + "epoch": 0.8721216858053268, + "grad_norm": 2.759496331490499, + "learning_rate": 1.689807459290431e-07, + "loss": 1.0352, + "step": 7253 + }, + { + "epoch": 0.8722419286959658, + "grad_norm": 3.3411013371554, + "learning_rate": 1.6866750759166437e-07, + "loss": 0.9486, + "step": 7254 + }, + { + "epoch": 0.8723621715866049, + "grad_norm": 2.6973108816804268, + "learning_rate": 1.6835454706596865e-07, + "loss": 1.0066, + "step": 7255 + }, + { + "epoch": 0.8724824144772441, + "grad_norm": 1.7601808995782542, + "learning_rate": 1.680418643994317e-07, + "loss": 0.9735, + "step": 7256 + }, + { + "epoch": 0.8726026573678831, + "grad_norm": 0.899771222866167, + "learning_rate": 1.6772945963948738e-07, + "loss": 0.8829, + "step": 7257 + }, + { + "epoch": 0.8727229002585222, + "grad_norm": 2.306240476676486, + "learning_rate": 1.6741733283352733e-07, + "loss": 1.0081, + "step": 7258 + }, + { + "epoch": 0.8728431431491613, + "grad_norm": 1.6886954135388155, + "learning_rate": 1.6710548402890102e-07, + "loss": 1.0767, + "step": 7259 + }, + { + "epoch": 0.8729633860398004, + "grad_norm": 1.7899369680505837, + "learning_rate": 1.6679391327291527e-07, + "loss": 0.9039, + "step": 7260 + }, + { + "epoch": 0.8730836289304394, + "grad_norm": 2.8617467163952206, + "learning_rate": 1.6648262061283492e-07, + "loss": 0.92, + "step": 7261 + }, + { + "epoch": 0.8732038718210786, + "grad_norm": 2.3786863529191833, + "learning_rate": 1.6617160609588353e-07, + "loss": 0.9694, + "step": 7262 + }, + { + "epoch": 0.8733241147117177, + "grad_norm": 2.090785076587432, + "learning_rate": 1.6586086976924163e-07, + "loss": 0.9514, + "step": 7263 + }, + { + "epoch": 0.8734443576023567, + "grad_norm": 1.8649018155963533, + "learning_rate": 1.6555041168004747e-07, + "loss": 1.0234, + "step": 7264 + }, + { + "epoch": 0.8735646004929959, + "grad_norm": 1.7580354676298018, + "learning_rate": 1.6524023187539715e-07, + "loss": 0.9257, + "step": 7265 + }, + { + "epoch": 0.873684843383635, + "grad_norm": 1.9653381575020976, + "learning_rate": 1.649303304023446e-07, + "loss": 0.9812, + "step": 7266 + }, + { + "epoch": 0.873805086274274, + "grad_norm": 2.3636400691250117, + "learning_rate": 1.6462070730790246e-07, + "loss": 1.0276, + "step": 7267 + }, + { + "epoch": 0.8739253291649132, + "grad_norm": 3.3088260645638523, + "learning_rate": 1.6431136263903912e-07, + "loss": 1.0269, + "step": 7268 + }, + { + "epoch": 0.8740455720555522, + "grad_norm": 2.1079415431015396, + "learning_rate": 1.6400229644268282e-07, + "loss": 0.9736, + "step": 7269 + }, + { + "epoch": 0.8741658149461913, + "grad_norm": 1.9175809946075622, + "learning_rate": 1.6369350876571852e-07, + "loss": 1.0469, + "step": 7270 + }, + { + "epoch": 0.8742860578368304, + "grad_norm": 3.2733771926313744, + "learning_rate": 1.6338499965498874e-07, + "loss": 1.0522, + "step": 7271 + }, + { + "epoch": 0.8744063007274695, + "grad_norm": 1.7897418202515547, + "learning_rate": 1.630767691572943e-07, + "loss": 1.0073, + "step": 7272 + }, + { + "epoch": 0.8745265436181086, + "grad_norm": 0.7530485423046167, + "learning_rate": 1.6276881731939306e-07, + "loss": 0.7724, + "step": 7273 + }, + { + "epoch": 0.8746467865087477, + "grad_norm": 1.9013321397683105, + "learning_rate": 1.6246114418800193e-07, + "loss": 0.9904, + "step": 7274 + }, + { + "epoch": 0.8747670293993868, + "grad_norm": 1.818882982563819, + "learning_rate": 1.6215374980979423e-07, + "loss": 1.0017, + "step": 7275 + }, + { + "epoch": 0.8748872722900258, + "grad_norm": 3.0721917455754664, + "learning_rate": 1.6184663423140133e-07, + "loss": 0.9241, + "step": 7276 + }, + { + "epoch": 0.875007515180665, + "grad_norm": 3.1391936879066367, + "learning_rate": 1.615397974994126e-07, + "loss": 0.8783, + "step": 7277 + }, + { + "epoch": 0.875127758071304, + "grad_norm": 1.6665232199291033, + "learning_rate": 1.6123323966037438e-07, + "loss": 1.0459, + "step": 7278 + }, + { + "epoch": 0.8752480009619431, + "grad_norm": 2.5226816279462976, + "learning_rate": 1.6092696076079216e-07, + "loss": 1.0209, + "step": 7279 + }, + { + "epoch": 0.8753682438525822, + "grad_norm": 2.1381921238804216, + "learning_rate": 1.6062096084712785e-07, + "loss": 0.974, + "step": 7280 + }, + { + "epoch": 0.8754884867432213, + "grad_norm": 1.8423437154215208, + "learning_rate": 1.6031523996580098e-07, + "loss": 0.942, + "step": 7281 + }, + { + "epoch": 0.8756087296338604, + "grad_norm": 2.0597667500930337, + "learning_rate": 1.6000979816318981e-07, + "loss": 0.8957, + "step": 7282 + }, + { + "epoch": 0.8757289725244994, + "grad_norm": 2.5175505666948337, + "learning_rate": 1.5970463548562886e-07, + "loss": 0.9882, + "step": 7283 + }, + { + "epoch": 0.8758492154151386, + "grad_norm": 1.6652788901234739, + "learning_rate": 1.5939975197941192e-07, + "loss": 0.9439, + "step": 7284 + }, + { + "epoch": 0.8759694583057777, + "grad_norm": 0.8331671827301587, + "learning_rate": 1.5909514769078892e-07, + "loss": 0.7708, + "step": 7285 + }, + { + "epoch": 0.8760897011964167, + "grad_norm": 1.9106155810562284, + "learning_rate": 1.5879082266596867e-07, + "loss": 1.0142, + "step": 7286 + }, + { + "epoch": 0.8762099440870559, + "grad_norm": 1.7932796167225125, + "learning_rate": 1.5848677695111645e-07, + "loss": 0.9578, + "step": 7287 + }, + { + "epoch": 0.8763301869776949, + "grad_norm": 2.842324706672195, + "learning_rate": 1.5818301059235562e-07, + "loss": 0.9344, + "step": 7288 + }, + { + "epoch": 0.876450429868334, + "grad_norm": 2.001986828031852, + "learning_rate": 1.578795236357684e-07, + "loss": 1.0502, + "step": 7289 + }, + { + "epoch": 0.8765706727589732, + "grad_norm": 2.571746704313047, + "learning_rate": 1.5757631612739218e-07, + "loss": 1.0917, + "step": 7290 + }, + { + "epoch": 0.8766909156496122, + "grad_norm": 0.7995129473546198, + "learning_rate": 1.572733881132242e-07, + "loss": 0.8516, + "step": 7291 + }, + { + "epoch": 0.8768111585402513, + "grad_norm": 0.7962500047309432, + "learning_rate": 1.5697073963921814e-07, + "loss": 0.8224, + "step": 7292 + }, + { + "epoch": 0.8769314014308904, + "grad_norm": 2.7072417382429426, + "learning_rate": 1.566683707512857e-07, + "loss": 1.0918, + "step": 7293 + }, + { + "epoch": 0.8770516443215295, + "grad_norm": 8.660953497076152, + "learning_rate": 1.5636628149529553e-07, + "loss": 1.0341, + "step": 7294 + }, + { + "epoch": 0.8771718872121685, + "grad_norm": 2.0905329343313217, + "learning_rate": 1.560644719170743e-07, + "loss": 1.0311, + "step": 7295 + }, + { + "epoch": 0.8772921301028077, + "grad_norm": 2.3670275542859707, + "learning_rate": 1.5576294206240692e-07, + "loss": 0.9566, + "step": 7296 + }, + { + "epoch": 0.8774123729934468, + "grad_norm": 6.126925730134803, + "learning_rate": 1.5546169197703507e-07, + "loss": 0.9131, + "step": 7297 + }, + { + "epoch": 0.8775326158840858, + "grad_norm": 2.5394494334078015, + "learning_rate": 1.5516072170665774e-07, + "loss": 1.0112, + "step": 7298 + }, + { + "epoch": 0.877652858774725, + "grad_norm": 1.9621416019783051, + "learning_rate": 1.5486003129693214e-07, + "loss": 1.1082, + "step": 7299 + }, + { + "epoch": 0.877773101665364, + "grad_norm": 2.5947914330783424, + "learning_rate": 1.545596207934725e-07, + "loss": 1.0167, + "step": 7300 + }, + { + "epoch": 0.8778933445560031, + "grad_norm": 1.8289497015092118, + "learning_rate": 1.5425949024185147e-07, + "loss": 1.0147, + "step": 7301 + }, + { + "epoch": 0.8780135874466423, + "grad_norm": 2.0170642534422765, + "learning_rate": 1.5395963968759818e-07, + "loss": 0.911, + "step": 7302 + }, + { + "epoch": 0.8781338303372813, + "grad_norm": 1.7512988311472668, + "learning_rate": 1.536600691761998e-07, + "loss": 0.8831, + "step": 7303 + }, + { + "epoch": 0.8782540732279204, + "grad_norm": 1.830050375615059, + "learning_rate": 1.5336077875310084e-07, + "loss": 0.9544, + "step": 7304 + }, + { + "epoch": 0.8783743161185595, + "grad_norm": 1.8941515355219771, + "learning_rate": 1.5306176846370321e-07, + "loss": 0.9762, + "step": 7305 + }, + { + "epoch": 0.8784945590091986, + "grad_norm": 2.4072406265436017, + "learning_rate": 1.5276303835336712e-07, + "loss": 0.976, + "step": 7306 + }, + { + "epoch": 0.8786148018998376, + "grad_norm": 0.7705699281590506, + "learning_rate": 1.524645884674094e-07, + "loss": 0.7721, + "step": 7307 + }, + { + "epoch": 0.8787350447904768, + "grad_norm": 2.170464476172757, + "learning_rate": 1.521664188511047e-07, + "loss": 1.029, + "step": 7308 + }, + { + "epoch": 0.8788552876811159, + "grad_norm": 2.042041225200285, + "learning_rate": 1.518685295496851e-07, + "loss": 1.0435, + "step": 7309 + }, + { + "epoch": 0.8789755305717549, + "grad_norm": 1.7848483765475442, + "learning_rate": 1.5157092060833975e-07, + "loss": 1.0919, + "step": 7310 + }, + { + "epoch": 0.879095773462394, + "grad_norm": 1.7077668654198608, + "learning_rate": 1.5127359207221658e-07, + "loss": 0.8962, + "step": 7311 + }, + { + "epoch": 0.8792160163530331, + "grad_norm": 1.9680559660699428, + "learning_rate": 1.5097654398641923e-07, + "loss": 0.974, + "step": 7312 + }, + { + "epoch": 0.8793362592436722, + "grad_norm": 1.4225647479508363, + "learning_rate": 1.5067977639601014e-07, + "loss": 0.9654, + "step": 7313 + }, + { + "epoch": 0.8794565021343113, + "grad_norm": 2.523312403173414, + "learning_rate": 1.5038328934600864e-07, + "loss": 0.9456, + "step": 7314 + }, + { + "epoch": 0.8795767450249504, + "grad_norm": 2.1310878389926686, + "learning_rate": 1.5008708288139161e-07, + "loss": 0.94, + "step": 7315 + }, + { + "epoch": 0.8796969879155895, + "grad_norm": 2.221620341755452, + "learning_rate": 1.497911570470931e-07, + "loss": 0.9702, + "step": 7316 + }, + { + "epoch": 0.8798172308062285, + "grad_norm": 1.8568211816610891, + "learning_rate": 1.494955118880048e-07, + "loss": 1.0953, + "step": 7317 + }, + { + "epoch": 0.8799374736968677, + "grad_norm": 2.0431845206397976, + "learning_rate": 1.4920014744897634e-07, + "loss": 0.9673, + "step": 7318 + }, + { + "epoch": 0.8800577165875068, + "grad_norm": 1.7990199679485257, + "learning_rate": 1.4890506377481392e-07, + "loss": 1.1026, + "step": 7319 + }, + { + "epoch": 0.8801779594781458, + "grad_norm": 1.5420827132379, + "learning_rate": 1.486102609102815e-07, + "loss": 0.8774, + "step": 7320 + }, + { + "epoch": 0.880298202368785, + "grad_norm": 2.844479142497278, + "learning_rate": 1.483157389001004e-07, + "loss": 1.0932, + "step": 7321 + }, + { + "epoch": 0.880418445259424, + "grad_norm": 2.4348156159322714, + "learning_rate": 1.4802149778894933e-07, + "loss": 1.028, + "step": 7322 + }, + { + "epoch": 0.8805386881500631, + "grad_norm": 1.7032281686875061, + "learning_rate": 1.4772753762146484e-07, + "loss": 1.111, + "step": 7323 + }, + { + "epoch": 0.8806589310407023, + "grad_norm": 1.86827700443524, + "learning_rate": 1.474338584422401e-07, + "loss": 0.9427, + "step": 7324 + }, + { + "epoch": 0.8807791739313413, + "grad_norm": 1.970290171945193, + "learning_rate": 1.4714046029582595e-07, + "loss": 0.9945, + "step": 7325 + }, + { + "epoch": 0.8808994168219804, + "grad_norm": 1.9220921463526235, + "learning_rate": 1.46847343226731e-07, + "loss": 0.996, + "step": 7326 + }, + { + "epoch": 0.8810196597126195, + "grad_norm": 1.8351350369776256, + "learning_rate": 1.465545072794203e-07, + "loss": 0.9307, + "step": 7327 + }, + { + "epoch": 0.8811399026032586, + "grad_norm": 1.6391746286270923, + "learning_rate": 1.4626195249831774e-07, + "loss": 0.9917, + "step": 7328 + }, + { + "epoch": 0.8812601454938976, + "grad_norm": 1.7679221814234618, + "learning_rate": 1.4596967892780244e-07, + "loss": 0.9593, + "step": 7329 + }, + { + "epoch": 0.8813803883845368, + "grad_norm": 1.9609607831932911, + "learning_rate": 1.4567768661221314e-07, + "loss": 0.9835, + "step": 7330 + }, + { + "epoch": 0.8815006312751759, + "grad_norm": 2.0253846215331106, + "learning_rate": 1.4538597559584442e-07, + "loss": 0.9823, + "step": 7331 + }, + { + "epoch": 0.8816208741658149, + "grad_norm": 2.0102830226402757, + "learning_rate": 1.4509454592294823e-07, + "loss": 1.0259, + "step": 7332 + }, + { + "epoch": 0.8817411170564541, + "grad_norm": 2.2734628787377495, + "learning_rate": 1.448033976377354e-07, + "loss": 1.0271, + "step": 7333 + }, + { + "epoch": 0.8818613599470931, + "grad_norm": 2.0867365189013216, + "learning_rate": 1.445125307843713e-07, + "loss": 0.9779, + "step": 7334 + }, + { + "epoch": 0.8819816028377322, + "grad_norm": 1.6002068713681072, + "learning_rate": 1.442219454069813e-07, + "loss": 0.9918, + "step": 7335 + }, + { + "epoch": 0.8821018457283714, + "grad_norm": 2.268363507771065, + "learning_rate": 1.4393164154964676e-07, + "loss": 0.8962, + "step": 7336 + }, + { + "epoch": 0.8822220886190104, + "grad_norm": 1.7641891447547395, + "learning_rate": 1.4364161925640649e-07, + "loss": 1.1807, + "step": 7337 + }, + { + "epoch": 0.8823423315096495, + "grad_norm": 1.8494345754585013, + "learning_rate": 1.4335187857125663e-07, + "loss": 1.0891, + "step": 7338 + }, + { + "epoch": 0.8824625744002886, + "grad_norm": 1.9802527676700128, + "learning_rate": 1.4306241953815023e-07, + "loss": 0.9912, + "step": 7339 + }, + { + "epoch": 0.8825828172909277, + "grad_norm": 1.6238374219329572, + "learning_rate": 1.4277324220099862e-07, + "loss": 0.9472, + "step": 7340 + }, + { + "epoch": 0.8827030601815667, + "grad_norm": 1.8408194058767662, + "learning_rate": 1.4248434660366938e-07, + "loss": 0.9778, + "step": 7341 + }, + { + "epoch": 0.8828233030722058, + "grad_norm": 2.4394961277734946, + "learning_rate": 1.4219573278998808e-07, + "loss": 0.9409, + "step": 7342 + }, + { + "epoch": 0.882943545962845, + "grad_norm": 2.406606276257273, + "learning_rate": 1.4190740080373685e-07, + "loss": 0.886, + "step": 7343 + }, + { + "epoch": 0.883063788853484, + "grad_norm": 2.2246638183362126, + "learning_rate": 1.4161935068865538e-07, + "loss": 1.0766, + "step": 7344 + }, + { + "epoch": 0.8831840317441231, + "grad_norm": 1.9792579094450602, + "learning_rate": 1.4133158248844113e-07, + "loss": 0.9986, + "step": 7345 + }, + { + "epoch": 0.8833042746347622, + "grad_norm": 1.829121701057883, + "learning_rate": 1.4104409624674785e-07, + "loss": 0.9731, + "step": 7346 + }, + { + "epoch": 0.8834245175254013, + "grad_norm": 2.064977833626584, + "learning_rate": 1.407568920071873e-07, + "loss": 1.0224, + "step": 7347 + }, + { + "epoch": 0.8835447604160404, + "grad_norm": 2.0596724114540357, + "learning_rate": 1.4046996981332782e-07, + "loss": 0.9215, + "step": 7348 + }, + { + "epoch": 0.8836650033066795, + "grad_norm": 1.9434513242812292, + "learning_rate": 1.4018332970869516e-07, + "loss": 1.0201, + "step": 7349 + }, + { + "epoch": 0.8837852461973186, + "grad_norm": 2.03630024978523, + "learning_rate": 1.398969717367733e-07, + "loss": 1.0898, + "step": 7350 + }, + { + "epoch": 0.8839054890879576, + "grad_norm": 2.2554719930849454, + "learning_rate": 1.396108959410014e-07, + "loss": 1.0002, + "step": 7351 + }, + { + "epoch": 0.8840257319785968, + "grad_norm": 1.5971218418514963, + "learning_rate": 1.3932510236477745e-07, + "loss": 1.0519, + "step": 7352 + }, + { + "epoch": 0.8841459748692359, + "grad_norm": 2.6976995982231635, + "learning_rate": 1.3903959105145636e-07, + "loss": 0.7978, + "step": 7353 + }, + { + "epoch": 0.8842662177598749, + "grad_norm": 2.035514228952854, + "learning_rate": 1.387543620443492e-07, + "loss": 1.0679, + "step": 7354 + }, + { + "epoch": 0.8843864606505141, + "grad_norm": 3.751580477492328, + "learning_rate": 1.3846941538672606e-07, + "loss": 1.079, + "step": 7355 + }, + { + "epoch": 0.8845067035411531, + "grad_norm": 2.3156913181543306, + "learning_rate": 1.3818475112181193e-07, + "loss": 1.0466, + "step": 7356 + }, + { + "epoch": 0.8846269464317922, + "grad_norm": 2.4261869376947494, + "learning_rate": 1.3790036929279091e-07, + "loss": 1.0308, + "step": 7357 + }, + { + "epoch": 0.8847471893224313, + "grad_norm": 2.387957512124517, + "learning_rate": 1.3761626994280363e-07, + "loss": 0.8264, + "step": 7358 + }, + { + "epoch": 0.8848674322130704, + "grad_norm": 1.9778670879563292, + "learning_rate": 1.3733245311494735e-07, + "loss": 0.9736, + "step": 7359 + }, + { + "epoch": 0.8849876751037095, + "grad_norm": 2.002242619122083, + "learning_rate": 1.3704891885227676e-07, + "loss": 0.9438, + "step": 7360 + }, + { + "epoch": 0.8851079179943486, + "grad_norm": 2.0438347288527874, + "learning_rate": 1.367656671978037e-07, + "loss": 1.0156, + "step": 7361 + }, + { + "epoch": 0.8852281608849877, + "grad_norm": 2.057745337060051, + "learning_rate": 1.36482698194498e-07, + "loss": 0.9746, + "step": 7362 + }, + { + "epoch": 0.8853484037756267, + "grad_norm": 1.9902556982573372, + "learning_rate": 1.3620001188528506e-07, + "loss": 0.9557, + "step": 7363 + }, + { + "epoch": 0.8854686466662659, + "grad_norm": 3.305853888105824, + "learning_rate": 1.3591760831304865e-07, + "loss": 0.9758, + "step": 7364 + }, + { + "epoch": 0.885588889556905, + "grad_norm": 1.785870765914378, + "learning_rate": 1.356354875206287e-07, + "loss": 1.0308, + "step": 7365 + }, + { + "epoch": 0.885709132447544, + "grad_norm": 1.9695358524473012, + "learning_rate": 1.3535364955082296e-07, + "loss": 0.9357, + "step": 7366 + }, + { + "epoch": 0.8858293753381832, + "grad_norm": 1.7780129985698911, + "learning_rate": 1.3507209444638613e-07, + "loss": 0.8833, + "step": 7367 + }, + { + "epoch": 0.8859496182288222, + "grad_norm": 1.8319871323729138, + "learning_rate": 1.347908222500298e-07, + "loss": 0.9745, + "step": 7368 + }, + { + "epoch": 0.8860698611194613, + "grad_norm": 1.9201907709833428, + "learning_rate": 1.3450983300442276e-07, + "loss": 0.9317, + "step": 7369 + }, + { + "epoch": 0.8861901040101005, + "grad_norm": 2.6982671051034584, + "learning_rate": 1.3422912675219068e-07, + "loss": 0.9724, + "step": 7370 + }, + { + "epoch": 0.8863103469007395, + "grad_norm": 1.5406677859170355, + "learning_rate": 1.339487035359166e-07, + "loss": 1.0304, + "step": 7371 + }, + { + "epoch": 0.8864305897913786, + "grad_norm": 1.5555660506681588, + "learning_rate": 1.336685633981409e-07, + "loss": 1.0833, + "step": 7372 + }, + { + "epoch": 0.8865508326820177, + "grad_norm": 1.9285428130837472, + "learning_rate": 1.333887063813597e-07, + "loss": 0.9804, + "step": 7373 + }, + { + "epoch": 0.8866710755726568, + "grad_norm": 1.8113704976713032, + "learning_rate": 1.331091325280278e-07, + "loss": 0.899, + "step": 7374 + }, + { + "epoch": 0.8867913184632958, + "grad_norm": 1.7526412514045417, + "learning_rate": 1.3282984188055625e-07, + "loss": 1.0256, + "step": 7375 + }, + { + "epoch": 0.8869115613539349, + "grad_norm": 2.147021350564324, + "learning_rate": 1.3255083448131288e-07, + "loss": 1.0313, + "step": 7376 + }, + { + "epoch": 0.8870318042445741, + "grad_norm": 2.105189487147044, + "learning_rate": 1.3227211037262365e-07, + "loss": 1.0256, + "step": 7377 + }, + { + "epoch": 0.8871520471352131, + "grad_norm": 2.66396799227528, + "learning_rate": 1.319936695967696e-07, + "loss": 1.0915, + "step": 7378 + }, + { + "epoch": 0.8872722900258522, + "grad_norm": 2.100569820297885, + "learning_rate": 1.3171551219599097e-07, + "loss": 1.0622, + "step": 7379 + }, + { + "epoch": 0.8873925329164913, + "grad_norm": 4.359566414298412, + "learning_rate": 1.3143763821248377e-07, + "loss": 1.018, + "step": 7380 + }, + { + "epoch": 0.8875127758071304, + "grad_norm": 1.961520833947447, + "learning_rate": 1.3116004768840118e-07, + "loss": 0.9566, + "step": 7381 + }, + { + "epoch": 0.8876330186977694, + "grad_norm": 1.6247609769653157, + "learning_rate": 1.3088274066585348e-07, + "loss": 0.9763, + "step": 7382 + }, + { + "epoch": 0.8877532615884086, + "grad_norm": 2.2418268957153633, + "learning_rate": 1.3060571718690749e-07, + "loss": 1.1396, + "step": 7383 + }, + { + "epoch": 0.8878735044790477, + "grad_norm": 0.7547806723756751, + "learning_rate": 1.3032897729358805e-07, + "loss": 0.8067, + "step": 7384 + }, + { + "epoch": 0.8879937473696867, + "grad_norm": 1.9530828852881335, + "learning_rate": 1.3005252102787645e-07, + "loss": 1.0375, + "step": 7385 + }, + { + "epoch": 0.8881139902603259, + "grad_norm": 1.6607696477097316, + "learning_rate": 1.297763484317105e-07, + "loss": 0.9694, + "step": 7386 + }, + { + "epoch": 0.888234233150965, + "grad_norm": 2.2231365415688025, + "learning_rate": 1.2950045954698551e-07, + "loss": 0.9432, + "step": 7387 + }, + { + "epoch": 0.888354476041604, + "grad_norm": 1.7649129596461557, + "learning_rate": 1.2922485441555343e-07, + "loss": 0.9925, + "step": 7388 + }, + { + "epoch": 0.8884747189322432, + "grad_norm": 2.02883039295223, + "learning_rate": 1.2894953307922363e-07, + "loss": 1.0538, + "step": 7389 + }, + { + "epoch": 0.8885949618228822, + "grad_norm": 2.1464586056801647, + "learning_rate": 1.2867449557976208e-07, + "loss": 1.0769, + "step": 7390 + }, + { + "epoch": 0.8887152047135213, + "grad_norm": 1.8894444081955355, + "learning_rate": 1.283997419588916e-07, + "loss": 0.9923, + "step": 7391 + }, + { + "epoch": 0.8888354476041604, + "grad_norm": 2.1455925988524114, + "learning_rate": 1.2812527225829216e-07, + "loss": 0.8531, + "step": 7392 + }, + { + "epoch": 0.8889556904947995, + "grad_norm": 2.352574431714455, + "learning_rate": 1.2785108651960052e-07, + "loss": 1.0027, + "step": 7393 + }, + { + "epoch": 0.8890759333854386, + "grad_norm": 2.204721075568773, + "learning_rate": 1.2757718478441094e-07, + "loss": 1.044, + "step": 7394 + }, + { + "epoch": 0.8891961762760777, + "grad_norm": 1.7270993049843266, + "learning_rate": 1.2730356709427302e-07, + "loss": 1.0085, + "step": 7395 + }, + { + "epoch": 0.8893164191667168, + "grad_norm": 1.5839996525301696, + "learning_rate": 1.2703023349069542e-07, + "loss": 0.8358, + "step": 7396 + }, + { + "epoch": 0.8894366620573558, + "grad_norm": 1.8263801659914425, + "learning_rate": 1.2675718401514223e-07, + "loss": 0.8547, + "step": 7397 + }, + { + "epoch": 0.889556904947995, + "grad_norm": 2.114781888713021, + "learning_rate": 1.264844187090346e-07, + "loss": 0.9837, + "step": 7398 + }, + { + "epoch": 0.889677147838634, + "grad_norm": 1.8654016230032473, + "learning_rate": 1.262119376137516e-07, + "loss": 0.9882, + "step": 7399 + }, + { + "epoch": 0.8897973907292731, + "grad_norm": 1.8079952955353558, + "learning_rate": 1.2593974077062707e-07, + "loss": 1.0866, + "step": 7400 + }, + { + "epoch": 0.8899176336199123, + "grad_norm": 1.6582558200043342, + "learning_rate": 1.2566782822095423e-07, + "loss": 0.872, + "step": 7401 + }, + { + "epoch": 0.8900378765105513, + "grad_norm": 3.390581543815342, + "learning_rate": 1.2539620000598162e-07, + "loss": 0.951, + "step": 7402 + }, + { + "epoch": 0.8901581194011904, + "grad_norm": 1.7900072959665967, + "learning_rate": 1.2512485616691492e-07, + "loss": 1.0356, + "step": 7403 + }, + { + "epoch": 0.8902783622918296, + "grad_norm": 1.4695404847426312, + "learning_rate": 1.2485379674491681e-07, + "loss": 1.0468, + "step": 7404 + }, + { + "epoch": 0.8903986051824686, + "grad_norm": 2.7106805948999484, + "learning_rate": 1.2458302178110657e-07, + "loss": 1.0256, + "step": 7405 + }, + { + "epoch": 0.8905188480731077, + "grad_norm": 1.9526488677455576, + "learning_rate": 1.2431253131656118e-07, + "loss": 1.0611, + "step": 7406 + }, + { + "epoch": 0.8906390909637467, + "grad_norm": 1.8667266787804995, + "learning_rate": 1.240423253923133e-07, + "loss": 1.0054, + "step": 7407 + }, + { + "epoch": 0.8907593338543859, + "grad_norm": 2.0716538024936684, + "learning_rate": 1.237724040493533e-07, + "loss": 0.9263, + "step": 7408 + }, + { + "epoch": 0.8908795767450249, + "grad_norm": 2.892079580180273, + "learning_rate": 1.2350276732862773e-07, + "loss": 0.9668, + "step": 7409 + }, + { + "epoch": 0.890999819635664, + "grad_norm": 0.8662905135045601, + "learning_rate": 1.2323341527103993e-07, + "loss": 0.8046, + "step": 7410 + }, + { + "epoch": 0.8911200625263032, + "grad_norm": 2.227313427031294, + "learning_rate": 1.2296434791745135e-07, + "loss": 1.0866, + "step": 7411 + }, + { + "epoch": 0.8912403054169422, + "grad_norm": 1.5987312412429033, + "learning_rate": 1.2269556530867875e-07, + "loss": 1.0028, + "step": 7412 + }, + { + "epoch": 0.8913605483075813, + "grad_norm": 2.233479246905196, + "learning_rate": 1.2242706748549614e-07, + "loss": 1.0551, + "step": 7413 + }, + { + "epoch": 0.8914807911982204, + "grad_norm": 2.027455938685965, + "learning_rate": 1.2215885448863473e-07, + "loss": 1.0576, + "step": 7414 + }, + { + "epoch": 0.8916010340888595, + "grad_norm": 1.868487747315536, + "learning_rate": 1.2189092635878152e-07, + "loss": 1.0347, + "step": 7415 + }, + { + "epoch": 0.8917212769794985, + "grad_norm": 2.2289997882115573, + "learning_rate": 1.216232831365822e-07, + "loss": 1.011, + "step": 7416 + }, + { + "epoch": 0.8918415198701377, + "grad_norm": 2.0204937705302903, + "learning_rate": 1.2135592486263678e-07, + "loss": 1.0467, + "step": 7417 + }, + { + "epoch": 0.8919617627607768, + "grad_norm": 1.8685782771487074, + "learning_rate": 1.2108885157750415e-07, + "loss": 0.8495, + "step": 7418 + }, + { + "epoch": 0.8920820056514158, + "grad_norm": 1.792743810036298, + "learning_rate": 1.2082206332169897e-07, + "loss": 1.04, + "step": 7419 + }, + { + "epoch": 0.892202248542055, + "grad_norm": 2.3778991070608098, + "learning_rate": 1.2055556013569225e-07, + "loss": 0.9701, + "step": 7420 + }, + { + "epoch": 0.892322491432694, + "grad_norm": 1.735740330414923, + "learning_rate": 1.2028934205991315e-07, + "loss": 1.0579, + "step": 7421 + }, + { + "epoch": 0.8924427343233331, + "grad_norm": 1.7562630160789496, + "learning_rate": 1.2002340913474607e-07, + "loss": 1.0025, + "step": 7422 + }, + { + "epoch": 0.8925629772139723, + "grad_norm": 2.2844931137505498, + "learning_rate": 1.1975776140053317e-07, + "loss": 0.9772, + "step": 7423 + }, + { + "epoch": 0.8926832201046113, + "grad_norm": 1.91670461976442, + "learning_rate": 1.194923988975729e-07, + "loss": 0.9703, + "step": 7424 + }, + { + "epoch": 0.8928034629952504, + "grad_norm": 2.205973197711939, + "learning_rate": 1.192273216661206e-07, + "loss": 0.9701, + "step": 7425 + }, + { + "epoch": 0.8929237058858895, + "grad_norm": 0.7813420050804886, + "learning_rate": 1.189625297463881e-07, + "loss": 0.8125, + "step": 7426 + }, + { + "epoch": 0.8930439487765286, + "grad_norm": 1.583069185341564, + "learning_rate": 1.1869802317854394e-07, + "loss": 1.0347, + "step": 7427 + }, + { + "epoch": 0.8931641916671677, + "grad_norm": 1.895787905753134, + "learning_rate": 1.1843380200271425e-07, + "loss": 0.9589, + "step": 7428 + }, + { + "epoch": 0.8932844345578068, + "grad_norm": 2.2254888790403835, + "learning_rate": 1.181698662589805e-07, + "loss": 1.0427, + "step": 7429 + }, + { + "epoch": 0.8934046774484459, + "grad_norm": 3.7824202058775236, + "learning_rate": 1.1790621598738249e-07, + "loss": 0.9951, + "step": 7430 + }, + { + "epoch": 0.8935249203390849, + "grad_norm": 2.0293624323957125, + "learning_rate": 1.1764285122791461e-07, + "loss": 0.9858, + "step": 7431 + }, + { + "epoch": 0.8936451632297241, + "grad_norm": 1.8313090969093262, + "learning_rate": 1.173797720205294e-07, + "loss": 1.0099, + "step": 7432 + }, + { + "epoch": 0.8937654061203631, + "grad_norm": 2.6012684882193025, + "learning_rate": 1.1711697840513602e-07, + "loss": 0.956, + "step": 7433 + }, + { + "epoch": 0.8938856490110022, + "grad_norm": 2.1439657386348183, + "learning_rate": 1.1685447042160012e-07, + "loss": 0.9477, + "step": 7434 + }, + { + "epoch": 0.8940058919016414, + "grad_norm": 1.5339916364101116, + "learning_rate": 1.1659224810974367e-07, + "loss": 0.9534, + "step": 7435 + }, + { + "epoch": 0.8941261347922804, + "grad_norm": 1.481309717318575, + "learning_rate": 1.1633031150934591e-07, + "loss": 0.9227, + "step": 7436 + }, + { + "epoch": 0.8942463776829195, + "grad_norm": 2.1059730506179797, + "learning_rate": 1.1606866066014176e-07, + "loss": 1.032, + "step": 7437 + }, + { + "epoch": 0.8943666205735585, + "grad_norm": 3.0486740669657286, + "learning_rate": 1.1580729560182434e-07, + "loss": 0.9933, + "step": 7438 + }, + { + "epoch": 0.8944868634641977, + "grad_norm": 2.1104129074479943, + "learning_rate": 1.1554621637404171e-07, + "loss": 0.9443, + "step": 7439 + }, + { + "epoch": 0.8946071063548368, + "grad_norm": 3.0592551655041094, + "learning_rate": 1.1528542301639999e-07, + "loss": 0.8476, + "step": 7440 + }, + { + "epoch": 0.8947273492454758, + "grad_norm": 2.353147359187068, + "learning_rate": 1.1502491556846105e-07, + "loss": 1.0635, + "step": 7441 + }, + { + "epoch": 0.894847592136115, + "grad_norm": 2.3569722857005133, + "learning_rate": 1.1476469406974331e-07, + "loss": 1.0555, + "step": 7442 + }, + { + "epoch": 0.894967835026754, + "grad_norm": 1.7497824806177849, + "learning_rate": 1.1450475855972341e-07, + "loss": 1.009, + "step": 7443 + }, + { + "epoch": 0.8950880779173931, + "grad_norm": 2.517188871491033, + "learning_rate": 1.1424510907783158e-07, + "loss": 0.9422, + "step": 7444 + }, + { + "epoch": 0.8952083208080323, + "grad_norm": 1.7181027667625073, + "learning_rate": 1.1398574566345787e-07, + "loss": 1.0675, + "step": 7445 + }, + { + "epoch": 0.8953285636986713, + "grad_norm": 2.3235826663733534, + "learning_rate": 1.1372666835594702e-07, + "loss": 1.0672, + "step": 7446 + }, + { + "epoch": 0.8954488065893104, + "grad_norm": 4.101837555881372, + "learning_rate": 1.1346787719460071e-07, + "loss": 0.957, + "step": 7447 + }, + { + "epoch": 0.8955690494799495, + "grad_norm": 1.9546302793133523, + "learning_rate": 1.1320937221867732e-07, + "loss": 0.9612, + "step": 7448 + }, + { + "epoch": 0.8956892923705886, + "grad_norm": 1.7304155673084707, + "learning_rate": 1.1295115346739192e-07, + "loss": 1.0349, + "step": 7449 + }, + { + "epoch": 0.8958095352612276, + "grad_norm": 3.55417181613157, + "learning_rate": 1.1269322097991629e-07, + "loss": 0.9642, + "step": 7450 + }, + { + "epoch": 0.8959297781518668, + "grad_norm": 2.275472038678967, + "learning_rate": 1.1243557479537846e-07, + "loss": 0.9183, + "step": 7451 + }, + { + "epoch": 0.8960500210425059, + "grad_norm": 2.057363066259203, + "learning_rate": 1.121782149528634e-07, + "loss": 0.9259, + "step": 7452 + }, + { + "epoch": 0.8961702639331449, + "grad_norm": 2.139018994732553, + "learning_rate": 1.1192114149141208e-07, + "loss": 1.0254, + "step": 7453 + }, + { + "epoch": 0.8962905068237841, + "grad_norm": 2.7791544702844693, + "learning_rate": 1.1166435445002197e-07, + "loss": 0.8879, + "step": 7454 + }, + { + "epoch": 0.8964107497144231, + "grad_norm": 2.1348502795857978, + "learning_rate": 1.1140785386764818e-07, + "loss": 0.9255, + "step": 7455 + }, + { + "epoch": 0.8965309926050622, + "grad_norm": 2.079621772673269, + "learning_rate": 1.1115163978320153e-07, + "loss": 0.9332, + "step": 7456 + }, + { + "epoch": 0.8966512354957014, + "grad_norm": 2.1425314996422404, + "learning_rate": 1.1089571223554917e-07, + "loss": 1.0628, + "step": 7457 + }, + { + "epoch": 0.8967714783863404, + "grad_norm": 3.3308595696940544, + "learning_rate": 1.1064007126351537e-07, + "loss": 1.0921, + "step": 7458 + }, + { + "epoch": 0.8968917212769795, + "grad_norm": 2.205998055658668, + "learning_rate": 1.1038471690588003e-07, + "loss": 1.0019, + "step": 7459 + }, + { + "epoch": 0.8970119641676186, + "grad_norm": 2.0136841885535253, + "learning_rate": 1.1012964920138145e-07, + "loss": 1.0405, + "step": 7460 + }, + { + "epoch": 0.8971322070582577, + "grad_norm": 1.6402450121475383, + "learning_rate": 1.0987486818871205e-07, + "loss": 0.9939, + "step": 7461 + }, + { + "epoch": 0.8972524499488967, + "grad_norm": 3.4357736587492598, + "learning_rate": 1.0962037390652245e-07, + "loss": 0.9682, + "step": 7462 + }, + { + "epoch": 0.8973726928395359, + "grad_norm": 1.9198112882817768, + "learning_rate": 1.0936616639341911e-07, + "loss": 0.956, + "step": 7463 + }, + { + "epoch": 0.897492935730175, + "grad_norm": 0.7525642190822828, + "learning_rate": 1.0911224568796473e-07, + "loss": 0.7843, + "step": 7464 + }, + { + "epoch": 0.897613178620814, + "grad_norm": 3.3451925713228228, + "learning_rate": 1.0885861182867984e-07, + "loss": 0.9464, + "step": 7465 + }, + { + "epoch": 0.8977334215114532, + "grad_norm": 1.8854289888713434, + "learning_rate": 1.0860526485403942e-07, + "loss": 0.9443, + "step": 7466 + }, + { + "epoch": 0.8978536644020922, + "grad_norm": 1.6413059819501963, + "learning_rate": 1.0835220480247675e-07, + "loss": 1.0094, + "step": 7467 + }, + { + "epoch": 0.8979739072927313, + "grad_norm": 2.5001398047457117, + "learning_rate": 1.0809943171238067e-07, + "loss": 1.0721, + "step": 7468 + }, + { + "epoch": 0.8980941501833704, + "grad_norm": 2.391227097827856, + "learning_rate": 1.078469456220965e-07, + "loss": 0.8653, + "step": 7469 + }, + { + "epoch": 0.8982143930740095, + "grad_norm": 1.7591174897669442, + "learning_rate": 1.0759474656992606e-07, + "loss": 0.9333, + "step": 7470 + }, + { + "epoch": 0.8983346359646486, + "grad_norm": 2.732692004383281, + "learning_rate": 1.0734283459412785e-07, + "loss": 1.0205, + "step": 7471 + }, + { + "epoch": 0.8984548788552876, + "grad_norm": 1.8673682344823386, + "learning_rate": 1.0709120973291707e-07, + "loss": 1.043, + "step": 7472 + }, + { + "epoch": 0.8985751217459268, + "grad_norm": 8.103992711989354, + "learning_rate": 1.0683987202446475e-07, + "loss": 1.0082, + "step": 7473 + }, + { + "epoch": 0.8986953646365659, + "grad_norm": 2.260981569737225, + "learning_rate": 1.0658882150689862e-07, + "loss": 0.9351, + "step": 7474 + }, + { + "epoch": 0.8988156075272049, + "grad_norm": 2.428693463761528, + "learning_rate": 1.0633805821830288e-07, + "loss": 1.0235, + "step": 7475 + }, + { + "epoch": 0.8989358504178441, + "grad_norm": 2.563747567520215, + "learning_rate": 1.0608758219671753e-07, + "loss": 1.0627, + "step": 7476 + }, + { + "epoch": 0.8990560933084831, + "grad_norm": 1.6529443249104496, + "learning_rate": 1.0583739348014065e-07, + "loss": 0.94, + "step": 7477 + }, + { + "epoch": 0.8991763361991222, + "grad_norm": 2.2029750042439726, + "learning_rate": 1.0558749210652518e-07, + "loss": 1.084, + "step": 7478 + }, + { + "epoch": 0.8992965790897613, + "grad_norm": 1.8571371508080983, + "learning_rate": 1.053378781137808e-07, + "loss": 1.0922, + "step": 7479 + }, + { + "epoch": 0.8994168219804004, + "grad_norm": 1.9584860505422184, + "learning_rate": 1.0508855153977392e-07, + "loss": 1.0147, + "step": 7480 + }, + { + "epoch": 0.8995370648710395, + "grad_norm": 3.0508078208870106, + "learning_rate": 1.0483951242232669e-07, + "loss": 0.9095, + "step": 7481 + }, + { + "epoch": 0.8996573077616786, + "grad_norm": 0.983866726882808, + "learning_rate": 1.0459076079921936e-07, + "loss": 0.8143, + "step": 7482 + }, + { + "epoch": 0.8997775506523177, + "grad_norm": 3.020427615226162, + "learning_rate": 1.0434229670818618e-07, + "loss": 1.0898, + "step": 7483 + }, + { + "epoch": 0.8998977935429567, + "grad_norm": 1.5061759170738382, + "learning_rate": 1.0409412018691944e-07, + "loss": 1.0348, + "step": 7484 + }, + { + "epoch": 0.9000180364335959, + "grad_norm": 2.066874979144322, + "learning_rate": 1.0384623127306724e-07, + "loss": 0.9878, + "step": 7485 + }, + { + "epoch": 0.900138279324235, + "grad_norm": 1.7843395009159015, + "learning_rate": 1.0359863000423397e-07, + "loss": 1.0311, + "step": 7486 + }, + { + "epoch": 0.900258522214874, + "grad_norm": 1.6899588296604346, + "learning_rate": 1.0335131641798112e-07, + "loss": 0.953, + "step": 7487 + }, + { + "epoch": 0.9003787651055132, + "grad_norm": 0.8133795086137875, + "learning_rate": 1.0310429055182512e-07, + "loss": 0.7932, + "step": 7488 + }, + { + "epoch": 0.9004990079961522, + "grad_norm": 1.669607175208239, + "learning_rate": 1.0285755244324024e-07, + "loss": 0.9777, + "step": 7489 + }, + { + "epoch": 0.9006192508867913, + "grad_norm": 1.5816741163495722, + "learning_rate": 1.0261110212965629e-07, + "loss": 0.9244, + "step": 7490 + }, + { + "epoch": 0.9007394937774305, + "grad_norm": 1.9464259503848125, + "learning_rate": 1.023649396484596e-07, + "loss": 1.0296, + "step": 7491 + }, + { + "epoch": 0.9008597366680695, + "grad_norm": 1.9075021852253287, + "learning_rate": 1.0211906503699275e-07, + "loss": 0.9117, + "step": 7492 + }, + { + "epoch": 0.9009799795587086, + "grad_norm": 2.88172226219168, + "learning_rate": 1.0187347833255455e-07, + "loss": 1.0546, + "step": 7493 + }, + { + "epoch": 0.9011002224493477, + "grad_norm": 1.9336145881531819, + "learning_rate": 1.0162817957240056e-07, + "loss": 1.0276, + "step": 7494 + }, + { + "epoch": 0.9012204653399868, + "grad_norm": 0.8845892724119152, + "learning_rate": 1.0138316879374253e-07, + "loss": 0.8675, + "step": 7495 + }, + { + "epoch": 0.9013407082306258, + "grad_norm": 2.2133693732732334, + "learning_rate": 1.0113844603374833e-07, + "loss": 0.9785, + "step": 7496 + }, + { + "epoch": 0.901460951121265, + "grad_norm": 2.1464687135105964, + "learning_rate": 1.0089401132954178e-07, + "loss": 0.9568, + "step": 7497 + }, + { + "epoch": 0.9015811940119041, + "grad_norm": 1.6424111951629476, + "learning_rate": 1.006498647182037e-07, + "loss": 0.9634, + "step": 7498 + }, + { + "epoch": 0.9017014369025431, + "grad_norm": 1.9918508806255562, + "learning_rate": 1.004060062367713e-07, + "loss": 0.9515, + "step": 7499 + }, + { + "epoch": 0.9018216797931822, + "grad_norm": 1.8854444161018689, + "learning_rate": 1.0016243592223728e-07, + "loss": 0.9308, + "step": 7500 + }, + { + "epoch": 0.9019419226838213, + "grad_norm": 1.8951797476598207, + "learning_rate": 9.991915381155114e-08, + "loss": 0.8943, + "step": 7501 + }, + { + "epoch": 0.9020621655744604, + "grad_norm": 2.123186571847154, + "learning_rate": 9.967615994161871e-08, + "loss": 0.9844, + "step": 7502 + }, + { + "epoch": 0.9021824084650995, + "grad_norm": 1.7665255790130447, + "learning_rate": 9.943345434930161e-08, + "loss": 1.0178, + "step": 7503 + }, + { + "epoch": 0.9023026513557386, + "grad_norm": 1.9218367285910878, + "learning_rate": 9.919103707141885e-08, + "loss": 0.9254, + "step": 7504 + }, + { + "epoch": 0.9024228942463777, + "grad_norm": 1.9453457136267895, + "learning_rate": 9.89489081447441e-08, + "loss": 1.0024, + "step": 7505 + }, + { + "epoch": 0.9025431371370167, + "grad_norm": 1.964841565838125, + "learning_rate": 9.870706760600844e-08, + "loss": 1.0648, + "step": 7506 + }, + { + "epoch": 0.9026633800276559, + "grad_norm": 1.8277781230780608, + "learning_rate": 9.846551549189918e-08, + "loss": 0.965, + "step": 7507 + }, + { + "epoch": 0.902783622918295, + "grad_norm": 2.4782987940146666, + "learning_rate": 9.822425183905902e-08, + "loss": 0.9269, + "step": 7508 + }, + { + "epoch": 0.902903865808934, + "grad_norm": 0.91943933097196, + "learning_rate": 9.798327668408823e-08, + "loss": 0.9889, + "step": 7509 + }, + { + "epoch": 0.9030241086995732, + "grad_norm": 2.4654861314391625, + "learning_rate": 9.774259006354158e-08, + "loss": 0.9258, + "step": 7510 + }, + { + "epoch": 0.9031443515902122, + "grad_norm": 2.0041481154674394, + "learning_rate": 9.750219201393184e-08, + "loss": 0.9994, + "step": 7511 + }, + { + "epoch": 0.9032645944808513, + "grad_norm": 1.8824248488624957, + "learning_rate": 9.726208257172697e-08, + "loss": 1.0144, + "step": 7512 + }, + { + "epoch": 0.9033848373714904, + "grad_norm": 2.35459749673762, + "learning_rate": 9.702226177335115e-08, + "loss": 0.9817, + "step": 7513 + }, + { + "epoch": 0.9035050802621295, + "grad_norm": 1.7217446334974578, + "learning_rate": 9.67827296551853e-08, + "loss": 0.9635, + "step": 7514 + }, + { + "epoch": 0.9036253231527686, + "grad_norm": 2.542382332358601, + "learning_rate": 9.65434862535659e-08, + "loss": 0.9209, + "step": 7515 + }, + { + "epoch": 0.9037455660434077, + "grad_norm": 2.7270779272792804, + "learning_rate": 9.630453160478635e-08, + "loss": 0.8842, + "step": 7516 + }, + { + "epoch": 0.9038658089340468, + "grad_norm": 1.594466927037951, + "learning_rate": 9.60658657450959e-08, + "loss": 1.0631, + "step": 7517 + }, + { + "epoch": 0.9039860518246858, + "grad_norm": 1.7610001885426478, + "learning_rate": 9.582748871069979e-08, + "loss": 1.0349, + "step": 7518 + }, + { + "epoch": 0.904106294715325, + "grad_norm": 1.8937957909212866, + "learning_rate": 9.558940053775954e-08, + "loss": 1.0682, + "step": 7519 + }, + { + "epoch": 0.904226537605964, + "grad_norm": 1.843278792346423, + "learning_rate": 9.535160126239294e-08, + "loss": 0.9155, + "step": 7520 + }, + { + "epoch": 0.9043467804966031, + "grad_norm": 1.6236612233714363, + "learning_rate": 9.511409092067424e-08, + "loss": 0.9461, + "step": 7521 + }, + { + "epoch": 0.9044670233872423, + "grad_norm": 2.025175666253921, + "learning_rate": 9.487686954863327e-08, + "loss": 0.9138, + "step": 7522 + }, + { + "epoch": 0.9045872662778813, + "grad_norm": 2.082163278712784, + "learning_rate": 9.46399371822566e-08, + "loss": 1.0088, + "step": 7523 + }, + { + "epoch": 0.9047075091685204, + "grad_norm": 2.166816363909786, + "learning_rate": 9.440329385748657e-08, + "loss": 0.9601, + "step": 7524 + }, + { + "epoch": 0.9048277520591596, + "grad_norm": 2.0043703252002145, + "learning_rate": 9.416693961022137e-08, + "loss": 0.9469, + "step": 7525 + }, + { + "epoch": 0.9049479949497986, + "grad_norm": 1.7824326989093746, + "learning_rate": 9.393087447631654e-08, + "loss": 1.0059, + "step": 7526 + }, + { + "epoch": 0.9050682378404377, + "grad_norm": 1.5576230040055976, + "learning_rate": 9.36950984915823e-08, + "loss": 0.9701, + "step": 7527 + }, + { + "epoch": 0.9051884807310768, + "grad_norm": 1.8805618446929864, + "learning_rate": 9.345961169178607e-08, + "loss": 0.9304, + "step": 7528 + }, + { + "epoch": 0.9053087236217159, + "grad_norm": 1.5933872632884751, + "learning_rate": 9.322441411265081e-08, + "loss": 0.9682, + "step": 7529 + }, + { + "epoch": 0.9054289665123549, + "grad_norm": 2.005217779228124, + "learning_rate": 9.298950578985554e-08, + "loss": 0.9706, + "step": 7530 + }, + { + "epoch": 0.905549209402994, + "grad_norm": 1.880332167783927, + "learning_rate": 9.275488675903665e-08, + "loss": 0.945, + "step": 7531 + }, + { + "epoch": 0.9056694522936332, + "grad_norm": 2.7662123579040343, + "learning_rate": 9.252055705578454e-08, + "loss": 0.9762, + "step": 7532 + }, + { + "epoch": 0.9057896951842722, + "grad_norm": 1.8801959205562897, + "learning_rate": 9.228651671564747e-08, + "loss": 0.9556, + "step": 7533 + }, + { + "epoch": 0.9059099380749113, + "grad_norm": 1.5315444040465043, + "learning_rate": 9.205276577412901e-08, + "loss": 1.0172, + "step": 7534 + }, + { + "epoch": 0.9060301809655504, + "grad_norm": 2.364412097080231, + "learning_rate": 9.181930426668905e-08, + "loss": 1.0013, + "step": 7535 + }, + { + "epoch": 0.9061504238561895, + "grad_norm": 1.9227610460697406, + "learning_rate": 9.158613222874346e-08, + "loss": 0.9167, + "step": 7536 + }, + { + "epoch": 0.9062706667468285, + "grad_norm": 1.6564749798829657, + "learning_rate": 9.135324969566394e-08, + "loss": 1.055, + "step": 7537 + }, + { + "epoch": 0.9063909096374677, + "grad_norm": 2.184580025717398, + "learning_rate": 9.112065670277913e-08, + "loss": 0.9923, + "step": 7538 + }, + { + "epoch": 0.9065111525281068, + "grad_norm": 1.9902440184115304, + "learning_rate": 9.088835328537303e-08, + "loss": 0.9665, + "step": 7539 + }, + { + "epoch": 0.9066313954187458, + "grad_norm": 2.221087232131331, + "learning_rate": 9.065633947868568e-08, + "loss": 0.9538, + "step": 7540 + }, + { + "epoch": 0.906751638309385, + "grad_norm": 3.581752753793405, + "learning_rate": 9.042461531791379e-08, + "loss": 1.0384, + "step": 7541 + }, + { + "epoch": 0.906871881200024, + "grad_norm": 1.877024257236933, + "learning_rate": 9.019318083820903e-08, + "loss": 1.0152, + "step": 7542 + }, + { + "epoch": 0.9069921240906631, + "grad_norm": 1.7710741066400744, + "learning_rate": 8.996203607468045e-08, + "loss": 1.0901, + "step": 7543 + }, + { + "epoch": 0.9071123669813023, + "grad_norm": 1.54516136734098, + "learning_rate": 8.973118106239241e-08, + "loss": 0.9922, + "step": 7544 + }, + { + "epoch": 0.9072326098719413, + "grad_norm": 2.0111374691148627, + "learning_rate": 8.95006158363656e-08, + "loss": 1.1838, + "step": 7545 + }, + { + "epoch": 0.9073528527625804, + "grad_norm": 2.1558739154516786, + "learning_rate": 8.9270340431576e-08, + "loss": 1.0116, + "step": 7546 + }, + { + "epoch": 0.9074730956532195, + "grad_norm": 2.0179349688967405, + "learning_rate": 8.904035488295658e-08, + "loss": 0.9749, + "step": 7547 + }, + { + "epoch": 0.9075933385438586, + "grad_norm": 0.6693680812574297, + "learning_rate": 8.881065922539632e-08, + "loss": 0.7697, + "step": 7548 + }, + { + "epoch": 0.9077135814344977, + "grad_norm": 1.7455537716339555, + "learning_rate": 8.85812534937389e-08, + "loss": 0.9719, + "step": 7549 + }, + { + "epoch": 0.9078338243251368, + "grad_norm": 2.908526769579464, + "learning_rate": 8.835213772278583e-08, + "loss": 0.9044, + "step": 7550 + }, + { + "epoch": 0.9079540672157759, + "grad_norm": 1.808716936633094, + "learning_rate": 8.812331194729373e-08, + "loss": 1.0285, + "step": 7551 + }, + { + "epoch": 0.9080743101064149, + "grad_norm": 1.9977617852802967, + "learning_rate": 8.789477620197461e-08, + "loss": 0.9561, + "step": 7552 + }, + { + "epoch": 0.9081945529970541, + "grad_norm": 2.3260136189344958, + "learning_rate": 8.766653052149831e-08, + "loss": 1.0304, + "step": 7553 + }, + { + "epoch": 0.9083147958876931, + "grad_norm": 2.1341919402989764, + "learning_rate": 8.743857494048823e-08, + "loss": 0.9794, + "step": 7554 + }, + { + "epoch": 0.9084350387783322, + "grad_norm": 1.9068221187354575, + "learning_rate": 8.721090949352605e-08, + "loss": 0.8648, + "step": 7555 + }, + { + "epoch": 0.9085552816689714, + "grad_norm": 2.8320682917836164, + "learning_rate": 8.698353421514793e-08, + "loss": 0.9671, + "step": 7556 + }, + { + "epoch": 0.9086755245596104, + "grad_norm": 2.147306609322124, + "learning_rate": 8.67564491398467e-08, + "loss": 1.0392, + "step": 7557 + }, + { + "epoch": 0.9087957674502495, + "grad_norm": 1.9335529978147163, + "learning_rate": 8.652965430207104e-08, + "loss": 0.975, + "step": 7558 + }, + { + "epoch": 0.9089160103408886, + "grad_norm": 2.0986131584628196, + "learning_rate": 8.630314973622521e-08, + "loss": 0.8909, + "step": 7559 + }, + { + "epoch": 0.9090362532315277, + "grad_norm": 2.0042210381865737, + "learning_rate": 8.607693547666995e-08, + "loss": 0.9468, + "step": 7560 + }, + { + "epoch": 0.9091564961221668, + "grad_norm": 0.8894145168122508, + "learning_rate": 8.585101155772201e-08, + "loss": 0.8169, + "step": 7561 + }, + { + "epoch": 0.9092767390128058, + "grad_norm": 1.8683957456270397, + "learning_rate": 8.562537801365377e-08, + "loss": 0.9212, + "step": 7562 + }, + { + "epoch": 0.909396981903445, + "grad_norm": 1.8338798517539106, + "learning_rate": 8.540003487869362e-08, + "loss": 0.9355, + "step": 7563 + }, + { + "epoch": 0.909517224794084, + "grad_norm": 1.8388522770986708, + "learning_rate": 8.517498218702557e-08, + "loss": 1.0321, + "step": 7564 + }, + { + "epoch": 0.9096374676847231, + "grad_norm": 1.9247441778247285, + "learning_rate": 8.49502199727905e-08, + "loss": 0.9344, + "step": 7565 + }, + { + "epoch": 0.9097577105753623, + "grad_norm": 2.276288633819975, + "learning_rate": 8.472574827008428e-08, + "loss": 0.8951, + "step": 7566 + }, + { + "epoch": 0.9098779534660013, + "grad_norm": 1.692296875072133, + "learning_rate": 8.450156711295942e-08, + "loss": 1.0752, + "step": 7567 + }, + { + "epoch": 0.9099981963566404, + "grad_norm": 2.2360433472132097, + "learning_rate": 8.427767653542383e-08, + "loss": 1.1064, + "step": 7568 + }, + { + "epoch": 0.9101184392472795, + "grad_norm": 1.989851475832813, + "learning_rate": 8.405407657144125e-08, + "loss": 0.9378, + "step": 7569 + }, + { + "epoch": 0.9102386821379186, + "grad_norm": 2.1729329777584208, + "learning_rate": 8.383076725493232e-08, + "loss": 0.9631, + "step": 7570 + }, + { + "epoch": 0.9103589250285576, + "grad_norm": 2.833854814298098, + "learning_rate": 8.360774861977216e-08, + "loss": 0.9178, + "step": 7571 + }, + { + "epoch": 0.9104791679191968, + "grad_norm": 2.076950774048886, + "learning_rate": 8.338502069979281e-08, + "loss": 0.9837, + "step": 7572 + }, + { + "epoch": 0.9105994108098359, + "grad_norm": 3.0755158030538525, + "learning_rate": 8.316258352878214e-08, + "loss": 1.034, + "step": 7573 + }, + { + "epoch": 0.9107196537004749, + "grad_norm": 1.8969390984000885, + "learning_rate": 8.294043714048338e-08, + "loss": 0.9477, + "step": 7574 + }, + { + "epoch": 0.9108398965911141, + "grad_norm": 0.7557456549526821, + "learning_rate": 8.271858156859624e-08, + "loss": 0.8427, + "step": 7575 + }, + { + "epoch": 0.9109601394817531, + "grad_norm": 1.712303417645246, + "learning_rate": 8.249701684677557e-08, + "loss": 0.9782, + "step": 7576 + }, + { + "epoch": 0.9110803823723922, + "grad_norm": 1.9101901373168364, + "learning_rate": 8.227574300863294e-08, + "loss": 1.0453, + "step": 7577 + }, + { + "epoch": 0.9112006252630314, + "grad_norm": 2.244922631168852, + "learning_rate": 8.205476008773548e-08, + "loss": 0.9293, + "step": 7578 + }, + { + "epoch": 0.9113208681536704, + "grad_norm": 2.0776289727932866, + "learning_rate": 8.183406811760596e-08, + "loss": 1.0621, + "step": 7579 + }, + { + "epoch": 0.9114411110443095, + "grad_norm": 1.4881630197255973, + "learning_rate": 8.161366713172313e-08, + "loss": 0.9792, + "step": 7580 + }, + { + "epoch": 0.9115613539349486, + "grad_norm": 3.4032506774170845, + "learning_rate": 8.139355716352137e-08, + "loss": 1.0813, + "step": 7581 + }, + { + "epoch": 0.9116815968255877, + "grad_norm": 1.6716191461463892, + "learning_rate": 8.117373824639196e-08, + "loss": 0.9376, + "step": 7582 + }, + { + "epoch": 0.9118018397162267, + "grad_norm": 0.7333750308348729, + "learning_rate": 8.095421041368067e-08, + "loss": 0.8298, + "step": 7583 + }, + { + "epoch": 0.9119220826068659, + "grad_norm": 1.9902692947094514, + "learning_rate": 8.073497369868999e-08, + "loss": 0.9393, + "step": 7584 + }, + { + "epoch": 0.912042325497505, + "grad_norm": 3.2414293988500646, + "learning_rate": 8.051602813467772e-08, + "loss": 0.9944, + "step": 7585 + }, + { + "epoch": 0.912162568388144, + "grad_norm": 1.9926460486364568, + "learning_rate": 8.029737375485756e-08, + "loss": 0.9488, + "step": 7586 + }, + { + "epoch": 0.9122828112787832, + "grad_norm": 2.0630127240600795, + "learning_rate": 8.007901059239986e-08, + "loss": 0.9644, + "step": 7587 + }, + { + "epoch": 0.9124030541694222, + "grad_norm": 1.6388719184269847, + "learning_rate": 7.986093868042964e-08, + "loss": 1.039, + "step": 7588 + }, + { + "epoch": 0.9125232970600613, + "grad_norm": 1.6589913623394104, + "learning_rate": 7.964315805202826e-08, + "loss": 0.9129, + "step": 7589 + }, + { + "epoch": 0.9126435399507005, + "grad_norm": 1.7860481821974623, + "learning_rate": 7.942566874023304e-08, + "loss": 0.9702, + "step": 7590 + }, + { + "epoch": 0.9127637828413395, + "grad_norm": 2.26336051636198, + "learning_rate": 7.920847077803649e-08, + "loss": 0.9332, + "step": 7591 + }, + { + "epoch": 0.9128840257319786, + "grad_norm": 1.9687833026309691, + "learning_rate": 7.899156419838826e-08, + "loss": 1.0591, + "step": 7592 + }, + { + "epoch": 0.9130042686226177, + "grad_norm": 1.9645332106193116, + "learning_rate": 7.87749490341918e-08, + "loss": 0.8929, + "step": 7593 + }, + { + "epoch": 0.9131245115132568, + "grad_norm": 2.3062010886528763, + "learning_rate": 7.855862531830836e-08, + "loss": 1.0701, + "step": 7594 + }, + { + "epoch": 0.9132447544038959, + "grad_norm": 1.9550576985423005, + "learning_rate": 7.834259308355373e-08, + "loss": 0.9662, + "step": 7595 + }, + { + "epoch": 0.9133649972945349, + "grad_norm": 3.3346173356576547, + "learning_rate": 7.812685236269989e-08, + "loss": 0.984, + "step": 7596 + }, + { + "epoch": 0.9134852401851741, + "grad_norm": 0.80348445255119, + "learning_rate": 7.791140318847445e-08, + "loss": 0.8239, + "step": 7597 + }, + { + "epoch": 0.9136054830758131, + "grad_norm": 1.5305381988922826, + "learning_rate": 7.769624559356081e-08, + "loss": 1.0409, + "step": 7598 + }, + { + "epoch": 0.9137257259664522, + "grad_norm": 2.7033914423701075, + "learning_rate": 7.748137961059842e-08, + "loss": 0.9948, + "step": 7599 + }, + { + "epoch": 0.9138459688570914, + "grad_norm": 2.2722052416779692, + "learning_rate": 7.726680527218211e-08, + "loss": 0.8957, + "step": 7600 + }, + { + "epoch": 0.9139662117477304, + "grad_norm": 1.70608040102975, + "learning_rate": 7.70525226108627e-08, + "loss": 0.9942, + "step": 7601 + }, + { + "epoch": 0.9140864546383695, + "grad_norm": 1.7670926096405197, + "learning_rate": 7.683853165914666e-08, + "loss": 1.0345, + "step": 7602 + }, + { + "epoch": 0.9142066975290086, + "grad_norm": 2.4337691095402305, + "learning_rate": 7.662483244949602e-08, + "loss": 1.0084, + "step": 7603 + }, + { + "epoch": 0.9143269404196477, + "grad_norm": 5.564373086746677, + "learning_rate": 7.641142501432951e-08, + "loss": 1.0402, + "step": 7604 + }, + { + "epoch": 0.9144471833102867, + "grad_norm": 1.8693369221228862, + "learning_rate": 7.619830938602013e-08, + "loss": 0.9756, + "step": 7605 + }, + { + "epoch": 0.9145674262009259, + "grad_norm": 1.8597742501539385, + "learning_rate": 7.598548559689777e-08, + "loss": 1.065, + "step": 7606 + }, + { + "epoch": 0.914687669091565, + "grad_norm": 2.2326924851706296, + "learning_rate": 7.577295367924751e-08, + "loss": 1.0478, + "step": 7607 + }, + { + "epoch": 0.914807911982204, + "grad_norm": 2.000260097757089, + "learning_rate": 7.556071366531002e-08, + "loss": 1.0611, + "step": 7608 + }, + { + "epoch": 0.9149281548728432, + "grad_norm": 2.188409016974142, + "learning_rate": 7.53487655872822e-08, + "loss": 1.0255, + "step": 7609 + }, + { + "epoch": 0.9150483977634822, + "grad_norm": 1.737638067344672, + "learning_rate": 7.513710947731656e-08, + "loss": 0.9763, + "step": 7610 + }, + { + "epoch": 0.9151686406541213, + "grad_norm": 2.1785268667340394, + "learning_rate": 7.492574536752095e-08, + "loss": 1.0878, + "step": 7611 + }, + { + "epoch": 0.9152888835447605, + "grad_norm": 1.8493614151414772, + "learning_rate": 7.471467328995907e-08, + "loss": 1.0208, + "step": 7612 + }, + { + "epoch": 0.9154091264353995, + "grad_norm": 8.932301109612963, + "learning_rate": 7.450389327665018e-08, + "loss": 0.8471, + "step": 7613 + }, + { + "epoch": 0.9155293693260386, + "grad_norm": 3.2750483443790595, + "learning_rate": 7.429340535957029e-08, + "loss": 0.9117, + "step": 7614 + }, + { + "epoch": 0.9156496122166777, + "grad_norm": 3.6183380208687193, + "learning_rate": 7.40832095706494e-08, + "loss": 0.9471, + "step": 7615 + }, + { + "epoch": 0.9157698551073168, + "grad_norm": 1.868792239320984, + "learning_rate": 7.387330594177443e-08, + "loss": 1.0385, + "step": 7616 + }, + { + "epoch": 0.9158900979979558, + "grad_norm": 2.393232830544297, + "learning_rate": 7.366369450478749e-08, + "loss": 1.029, + "step": 7617 + }, + { + "epoch": 0.916010340888595, + "grad_norm": 2.1672567762153876, + "learning_rate": 7.345437529148646e-08, + "loss": 0.8977, + "step": 7618 + }, + { + "epoch": 0.9161305837792341, + "grad_norm": 2.059100499476254, + "learning_rate": 7.324534833362483e-08, + "loss": 0.9617, + "step": 7619 + }, + { + "epoch": 0.9162508266698731, + "grad_norm": 1.8255017361660102, + "learning_rate": 7.303661366291192e-08, + "loss": 0.9217, + "step": 7620 + }, + { + "epoch": 0.9163710695605123, + "grad_norm": 1.955515565818531, + "learning_rate": 7.28281713110126e-08, + "loss": 1.0503, + "step": 7621 + }, + { + "epoch": 0.9164913124511513, + "grad_norm": 2.27074795017068, + "learning_rate": 7.262002130954759e-08, + "loss": 1.0127, + "step": 7622 + }, + { + "epoch": 0.9166115553417904, + "grad_norm": 1.7755114825819274, + "learning_rate": 7.241216369009296e-08, + "loss": 1.026, + "step": 7623 + }, + { + "epoch": 0.9167317982324296, + "grad_norm": 1.7901324275381918, + "learning_rate": 7.220459848418037e-08, + "loss": 0.8999, + "step": 7624 + }, + { + "epoch": 0.9168520411230686, + "grad_norm": 1.848713997470336, + "learning_rate": 7.199732572329708e-08, + "loss": 1.0367, + "step": 7625 + }, + { + "epoch": 0.9169722840137077, + "grad_norm": 2.1993282419596554, + "learning_rate": 7.179034543888684e-08, + "loss": 0.9981, + "step": 7626 + }, + { + "epoch": 0.9170925269043467, + "grad_norm": 3.3081429094398787, + "learning_rate": 7.158365766234808e-08, + "loss": 1.0091, + "step": 7627 + }, + { + "epoch": 0.9172127697949859, + "grad_norm": 1.897932825393088, + "learning_rate": 7.137726242503527e-08, + "loss": 0.9647, + "step": 7628 + }, + { + "epoch": 0.917333012685625, + "grad_norm": 2.3254314432481804, + "learning_rate": 7.11711597582585e-08, + "loss": 1.022, + "step": 7629 + }, + { + "epoch": 0.917453255576264, + "grad_norm": 2.039832309674748, + "learning_rate": 7.096534969328271e-08, + "loss": 1.0341, + "step": 7630 + }, + { + "epoch": 0.9175734984669032, + "grad_norm": 2.080700074821233, + "learning_rate": 7.075983226132987e-08, + "loss": 1.0794, + "step": 7631 + }, + { + "epoch": 0.9176937413575422, + "grad_norm": 3.4221385000840225, + "learning_rate": 7.055460749357656e-08, + "loss": 1.0383, + "step": 7632 + }, + { + "epoch": 0.9178139842481813, + "grad_norm": 3.114561423200178, + "learning_rate": 7.034967542115521e-08, + "loss": 0.9405, + "step": 7633 + }, + { + "epoch": 0.9179342271388204, + "grad_norm": 2.3908400096071016, + "learning_rate": 7.014503607515388e-08, + "loss": 0.9877, + "step": 7634 + }, + { + "epoch": 0.9180544700294595, + "grad_norm": 2.084314293384853, + "learning_rate": 6.994068948661592e-08, + "loss": 0.9195, + "step": 7635 + }, + { + "epoch": 0.9181747129200986, + "grad_norm": 2.3186316776161995, + "learning_rate": 6.973663568654142e-08, + "loss": 1.0043, + "step": 7636 + }, + { + "epoch": 0.9182949558107377, + "grad_norm": 2.106190630248664, + "learning_rate": 6.953287470588386e-08, + "loss": 0.8918, + "step": 7637 + }, + { + "epoch": 0.9184151987013768, + "grad_norm": 2.3027469029623693, + "learning_rate": 6.932940657555452e-08, + "loss": 1.0961, + "step": 7638 + }, + { + "epoch": 0.9185354415920158, + "grad_norm": 1.4211598368876113, + "learning_rate": 6.912623132641938e-08, + "loss": 1.0001, + "step": 7639 + }, + { + "epoch": 0.918655684482655, + "grad_norm": 3.216307370637482, + "learning_rate": 6.892334898929952e-08, + "loss": 1.0045, + "step": 7640 + }, + { + "epoch": 0.918775927373294, + "grad_norm": 1.9800211082643446, + "learning_rate": 6.872075959497236e-08, + "loss": 1.0821, + "step": 7641 + }, + { + "epoch": 0.9188961702639331, + "grad_norm": 1.8414172917109022, + "learning_rate": 6.85184631741702e-08, + "loss": 1.0617, + "step": 7642 + }, + { + "epoch": 0.9190164131545723, + "grad_norm": 2.222739595214875, + "learning_rate": 6.831645975758161e-08, + "loss": 1.0103, + "step": 7643 + }, + { + "epoch": 0.9191366560452113, + "grad_norm": 2.1611763889006492, + "learning_rate": 6.811474937585026e-08, + "loss": 0.907, + "step": 7644 + }, + { + "epoch": 0.9192568989358504, + "grad_norm": 1.7368315748871985, + "learning_rate": 6.79133320595755e-08, + "loss": 1.0299, + "step": 7645 + }, + { + "epoch": 0.9193771418264896, + "grad_norm": 2.1024060453089946, + "learning_rate": 6.771220783931198e-08, + "loss": 0.9894, + "step": 7646 + }, + { + "epoch": 0.9194973847171286, + "grad_norm": 0.8485620818166945, + "learning_rate": 6.751137674556994e-08, + "loss": 0.8826, + "step": 7647 + }, + { + "epoch": 0.9196176276077677, + "grad_norm": 2.2084141002736226, + "learning_rate": 6.731083880881572e-08, + "loss": 1.0065, + "step": 7648 + }, + { + "epoch": 0.9197378704984068, + "grad_norm": 2.7235282913455783, + "learning_rate": 6.711059405947072e-08, + "loss": 1.0503, + "step": 7649 + }, + { + "epoch": 0.9198581133890459, + "grad_norm": 1.9404908752777306, + "learning_rate": 6.691064252791156e-08, + "loss": 1.0101, + "step": 7650 + }, + { + "epoch": 0.9199783562796849, + "grad_norm": 1.5769047315578921, + "learning_rate": 6.67109842444713e-08, + "loss": 1.0121, + "step": 7651 + }, + { + "epoch": 0.9200985991703241, + "grad_norm": 1.8342960893076972, + "learning_rate": 6.651161923943704e-08, + "loss": 1.0016, + "step": 7652 + }, + { + "epoch": 0.9202188420609632, + "grad_norm": 2.0055904932521833, + "learning_rate": 6.631254754305326e-08, + "loss": 1.0066, + "step": 7653 + }, + { + "epoch": 0.9203390849516022, + "grad_norm": 2.6016629403002303, + "learning_rate": 6.611376918551848e-08, + "loss": 1.0184, + "step": 7654 + }, + { + "epoch": 0.9204593278422414, + "grad_norm": 2.2612020531542973, + "learning_rate": 6.591528419698744e-08, + "loss": 1.0309, + "step": 7655 + }, + { + "epoch": 0.9205795707328804, + "grad_norm": 2.4401637463266104, + "learning_rate": 6.571709260756986e-08, + "loss": 1.0683, + "step": 7656 + }, + { + "epoch": 0.9206998136235195, + "grad_norm": 2.2324383218581603, + "learning_rate": 6.551919444733122e-08, + "loss": 1.0006, + "step": 7657 + }, + { + "epoch": 0.9208200565141585, + "grad_norm": 1.8913132857016657, + "learning_rate": 6.53215897462931e-08, + "loss": 0.8964, + "step": 7658 + }, + { + "epoch": 0.9209402994047977, + "grad_norm": 2.0450959372926034, + "learning_rate": 6.512427853443103e-08, + "loss": 0.9903, + "step": 7659 + }, + { + "epoch": 0.9210605422954368, + "grad_norm": 1.7879653298304359, + "learning_rate": 6.492726084167799e-08, + "loss": 0.9951, + "step": 7660 + }, + { + "epoch": 0.9211807851860758, + "grad_norm": 0.7889747570844309, + "learning_rate": 6.473053669792072e-08, + "loss": 0.8119, + "step": 7661 + }, + { + "epoch": 0.921301028076715, + "grad_norm": 2.089607924230777, + "learning_rate": 6.453410613300248e-08, + "loss": 0.9694, + "step": 7662 + }, + { + "epoch": 0.921421270967354, + "grad_norm": 1.8001649675008358, + "learning_rate": 6.43379691767214e-08, + "loss": 0.8216, + "step": 7663 + }, + { + "epoch": 0.9215415138579931, + "grad_norm": 0.7273755600280232, + "learning_rate": 6.414212585883105e-08, + "loss": 0.7884, + "step": 7664 + }, + { + "epoch": 0.9216617567486323, + "grad_norm": 2.264105872876746, + "learning_rate": 6.394657620904143e-08, + "loss": 0.9346, + "step": 7665 + }, + { + "epoch": 0.9217819996392713, + "grad_norm": 1.843922364534875, + "learning_rate": 6.375132025701657e-08, + "loss": 0.9538, + "step": 7666 + }, + { + "epoch": 0.9219022425299104, + "grad_norm": 4.130768874599658, + "learning_rate": 6.355635803237724e-08, + "loss": 0.9288, + "step": 7667 + }, + { + "epoch": 0.9220224854205495, + "grad_norm": 2.2456478837352507, + "learning_rate": 6.336168956469867e-08, + "loss": 1.0341, + "step": 7668 + }, + { + "epoch": 0.9221427283111886, + "grad_norm": 1.7633843171669974, + "learning_rate": 6.316731488351168e-08, + "loss": 0.9534, + "step": 7669 + }, + { + "epoch": 0.9222629712018277, + "grad_norm": 1.997827661445713, + "learning_rate": 6.297323401830334e-08, + "loss": 0.8671, + "step": 7670 + }, + { + "epoch": 0.9223832140924668, + "grad_norm": 2.2899834664759915, + "learning_rate": 6.277944699851523e-08, + "loss": 0.9261, + "step": 7671 + }, + { + "epoch": 0.9225034569831059, + "grad_norm": 1.9529247944264385, + "learning_rate": 6.25859538535447e-08, + "loss": 0.9703, + "step": 7672 + }, + { + "epoch": 0.9226236998737449, + "grad_norm": 4.314522089478243, + "learning_rate": 6.239275461274474e-08, + "loss": 1.0162, + "step": 7673 + }, + { + "epoch": 0.9227439427643841, + "grad_norm": 2.005238229768169, + "learning_rate": 6.219984930542299e-08, + "loss": 1.0936, + "step": 7674 + }, + { + "epoch": 0.9228641856550232, + "grad_norm": 1.9563588949979493, + "learning_rate": 6.200723796084383e-08, + "loss": 0.9957, + "step": 7675 + }, + { + "epoch": 0.9229844285456622, + "grad_norm": 0.7913196120895009, + "learning_rate": 6.181492060822546e-08, + "loss": 0.8676, + "step": 7676 + }, + { + "epoch": 0.9231046714363014, + "grad_norm": 2.0206258083899993, + "learning_rate": 6.162289727674274e-08, + "loss": 1.0543, + "step": 7677 + }, + { + "epoch": 0.9232249143269404, + "grad_norm": 2.386131101896334, + "learning_rate": 6.143116799552527e-08, + "loss": 1.1148, + "step": 7678 + }, + { + "epoch": 0.9233451572175795, + "grad_norm": 2.3826209819934534, + "learning_rate": 6.123973279365802e-08, + "loss": 0.7982, + "step": 7679 + }, + { + "epoch": 0.9234654001082186, + "grad_norm": 1.8464609679273216, + "learning_rate": 6.10485917001824e-08, + "loss": 1.0147, + "step": 7680 + }, + { + "epoch": 0.9235856429988577, + "grad_norm": 1.89200245513918, + "learning_rate": 6.085774474409322e-08, + "loss": 1.0467, + "step": 7681 + }, + { + "epoch": 0.9237058858894968, + "grad_norm": 2.296251381015518, + "learning_rate": 6.066719195434267e-08, + "loss": 0.9384, + "step": 7682 + }, + { + "epoch": 0.9238261287801359, + "grad_norm": 2.1477125019631624, + "learning_rate": 6.047693335983717e-08, + "loss": 0.9018, + "step": 7683 + }, + { + "epoch": 0.923946371670775, + "grad_norm": 2.741309653075367, + "learning_rate": 6.028696898943853e-08, + "loss": 1.0551, + "step": 7684 + }, + { + "epoch": 0.924066614561414, + "grad_norm": 2.0699227596006793, + "learning_rate": 6.00972988719648e-08, + "loss": 0.9423, + "step": 7685 + }, + { + "epoch": 0.9241868574520532, + "grad_norm": 3.7999993173699016, + "learning_rate": 5.990792303618807e-08, + "loss": 0.9404, + "step": 7686 + }, + { + "epoch": 0.9243071003426923, + "grad_norm": 1.762234157777325, + "learning_rate": 5.971884151083695e-08, + "loss": 0.9296, + "step": 7687 + }, + { + "epoch": 0.9244273432333313, + "grad_norm": 1.8397158438571202, + "learning_rate": 5.9530054324595124e-08, + "loss": 0.9816, + "step": 7688 + }, + { + "epoch": 0.9245475861239704, + "grad_norm": 0.7387611000401867, + "learning_rate": 5.934156150610103e-08, + "loss": 0.813, + "step": 7689 + }, + { + "epoch": 0.9246678290146095, + "grad_norm": 2.2812675579910784, + "learning_rate": 5.915336308394914e-08, + "loss": 1.0297, + "step": 7690 + }, + { + "epoch": 0.9247880719052486, + "grad_norm": 1.6825808423191235, + "learning_rate": 5.89654590866886e-08, + "loss": 1.0099, + "step": 7691 + }, + { + "epoch": 0.9249083147958876, + "grad_norm": 2.1875459938663604, + "learning_rate": 5.877784954282483e-08, + "loss": 1.1206, + "step": 7692 + }, + { + "epoch": 0.9250285576865268, + "grad_norm": 1.7955459323447875, + "learning_rate": 5.8590534480817963e-08, + "loss": 0.9634, + "step": 7693 + }, + { + "epoch": 0.9251488005771659, + "grad_norm": 2.6293675554993907, + "learning_rate": 5.840351392908349e-08, + "loss": 0.9596, + "step": 7694 + }, + { + "epoch": 0.9252690434678049, + "grad_norm": 2.487104056586164, + "learning_rate": 5.821678791599205e-08, + "loss": 0.9413, + "step": 7695 + }, + { + "epoch": 0.9253892863584441, + "grad_norm": 2.019648830746606, + "learning_rate": 5.803035646986965e-08, + "loss": 1.0441, + "step": 7696 + }, + { + "epoch": 0.9255095292490831, + "grad_norm": 2.5002267734672, + "learning_rate": 5.7844219618998766e-08, + "loss": 0.9106, + "step": 7697 + }, + { + "epoch": 0.9256297721397222, + "grad_norm": 2.156228217415734, + "learning_rate": 5.765837739161505e-08, + "loss": 0.9492, + "step": 7698 + }, + { + "epoch": 0.9257500150303614, + "grad_norm": 1.6478704408253986, + "learning_rate": 5.7472829815911504e-08, + "loss": 0.9809, + "step": 7699 + }, + { + "epoch": 0.9258702579210004, + "grad_norm": 1.8252594489049085, + "learning_rate": 5.7287576920035164e-08, + "loss": 1.0525, + "step": 7700 + }, + { + "epoch": 0.9259905008116395, + "grad_norm": 1.9676124071440784, + "learning_rate": 5.7102618732088435e-08, + "loss": 1.0029, + "step": 7701 + }, + { + "epoch": 0.9261107437022786, + "grad_norm": 1.7042518177848165, + "learning_rate": 5.6917955280130216e-08, + "loss": 0.982, + "step": 7702 + }, + { + "epoch": 0.9262309865929177, + "grad_norm": 2.2859894105827854, + "learning_rate": 5.6733586592172755e-08, + "loss": 0.9537, + "step": 7703 + }, + { + "epoch": 0.9263512294835567, + "grad_norm": 1.985853710552919, + "learning_rate": 5.6549512696185244e-08, + "loss": 1.0383, + "step": 7704 + }, + { + "epoch": 0.9264714723741959, + "grad_norm": 1.9231608981357768, + "learning_rate": 5.636573362009156e-08, + "loss": 0.9205, + "step": 7705 + }, + { + "epoch": 0.926591715264835, + "grad_norm": 2.10310142923328, + "learning_rate": 5.618224939177074e-08, + "loss": 1.0056, + "step": 7706 + }, + { + "epoch": 0.926711958155474, + "grad_norm": 1.8442751249794707, + "learning_rate": 5.599906003905719e-08, + "loss": 0.9402, + "step": 7707 + }, + { + "epoch": 0.9268322010461132, + "grad_norm": 2.322195297472474, + "learning_rate": 5.581616558974023e-08, + "loss": 1.0572, + "step": 7708 + }, + { + "epoch": 0.9269524439367522, + "grad_norm": 1.8611767798394645, + "learning_rate": 5.5633566071565444e-08, + "loss": 1.0236, + "step": 7709 + }, + { + "epoch": 0.9270726868273913, + "grad_norm": 2.0178378474377583, + "learning_rate": 5.5451261512232896e-08, + "loss": 0.944, + "step": 7710 + }, + { + "epoch": 0.9271929297180305, + "grad_norm": 1.9369215563249962, + "learning_rate": 5.5269251939397576e-08, + "loss": 0.8602, + "step": 7711 + }, + { + "epoch": 0.9273131726086695, + "grad_norm": 1.992108990487335, + "learning_rate": 5.508753738067073e-08, + "loss": 1.0038, + "step": 7712 + }, + { + "epoch": 0.9274334154993086, + "grad_norm": 3.0219913937580554, + "learning_rate": 5.4906117863617875e-08, + "loss": 1.0313, + "step": 7713 + }, + { + "epoch": 0.9275536583899477, + "grad_norm": 1.9405330175447664, + "learning_rate": 5.4724993415760533e-08, + "loss": 1.0162, + "step": 7714 + }, + { + "epoch": 0.9276739012805868, + "grad_norm": 2.191481862659149, + "learning_rate": 5.454416406457496e-08, + "loss": 0.9826, + "step": 7715 + }, + { + "epoch": 0.9277941441712259, + "grad_norm": 2.5239256395549146, + "learning_rate": 5.436362983749299e-08, + "loss": 0.9742, + "step": 7716 + }, + { + "epoch": 0.927914387061865, + "grad_norm": 2.0145787325480753, + "learning_rate": 5.418339076190137e-08, + "loss": 0.8823, + "step": 7717 + }, + { + "epoch": 0.9280346299525041, + "grad_norm": 1.9225230171204797, + "learning_rate": 5.400344686514202e-08, + "loss": 1.1239, + "step": 7718 + }, + { + "epoch": 0.9281548728431431, + "grad_norm": 2.014227804261707, + "learning_rate": 5.38237981745131e-08, + "loss": 0.9016, + "step": 7719 + }, + { + "epoch": 0.9282751157337822, + "grad_norm": 1.7619681509562342, + "learning_rate": 5.364444471726592e-08, + "loss": 1.0516, + "step": 7720 + }, + { + "epoch": 0.9283953586244214, + "grad_norm": 1.9751266873445843, + "learning_rate": 5.346538652060939e-08, + "loss": 1.0383, + "step": 7721 + }, + { + "epoch": 0.9285156015150604, + "grad_norm": 2.3547235578523926, + "learning_rate": 5.3286623611705994e-08, + "loss": 0.9408, + "step": 7722 + }, + { + "epoch": 0.9286358444056995, + "grad_norm": 0.8242943534707659, + "learning_rate": 5.3108156017673824e-08, + "loss": 0.8431, + "step": 7723 + }, + { + "epoch": 0.9287560872963386, + "grad_norm": 1.7900022345800677, + "learning_rate": 5.2929983765586775e-08, + "loss": 0.9529, + "step": 7724 + }, + { + "epoch": 0.9288763301869777, + "grad_norm": 1.7351582451147214, + "learning_rate": 5.275210688247278e-08, + "loss": 0.8633, + "step": 7725 + }, + { + "epoch": 0.9289965730776167, + "grad_norm": 3.918107248066167, + "learning_rate": 5.257452539531604e-08, + "loss": 1.0898, + "step": 7726 + }, + { + "epoch": 0.9291168159682559, + "grad_norm": 1.6806659776054966, + "learning_rate": 5.2397239331055445e-08, + "loss": 0.9203, + "step": 7727 + }, + { + "epoch": 0.929237058858895, + "grad_norm": 2.40020925086481, + "learning_rate": 5.2220248716585036e-08, + "loss": 1.0529, + "step": 7728 + }, + { + "epoch": 0.929357301749534, + "grad_norm": 2.3232948326120693, + "learning_rate": 5.204355357875445e-08, + "loss": 0.9905, + "step": 7729 + }, + { + "epoch": 0.9294775446401732, + "grad_norm": 2.1668786409807934, + "learning_rate": 5.1867153944367584e-08, + "loss": 0.9406, + "step": 7730 + }, + { + "epoch": 0.9295977875308122, + "grad_norm": 1.6071595433292345, + "learning_rate": 5.16910498401848e-08, + "loss": 0.9716, + "step": 7731 + }, + { + "epoch": 0.9297180304214513, + "grad_norm": 2.468457361574359, + "learning_rate": 5.151524129292073e-08, + "loss": 1.0728, + "step": 7732 + }, + { + "epoch": 0.9298382733120905, + "grad_norm": 2.0800727378524724, + "learning_rate": 5.1339728329245155e-08, + "loss": 0.9021, + "step": 7733 + }, + { + "epoch": 0.9299585162027295, + "grad_norm": 2.1933772017585884, + "learning_rate": 5.116451097578367e-08, + "loss": 1.034, + "step": 7734 + }, + { + "epoch": 0.9300787590933686, + "grad_norm": 1.8580310356126797, + "learning_rate": 5.0989589259115895e-08, + "loss": 0.9804, + "step": 7735 + }, + { + "epoch": 0.9301990019840077, + "grad_norm": 3.265758748943165, + "learning_rate": 5.081496320577816e-08, + "loss": 0.9524, + "step": 7736 + }, + { + "epoch": 0.9303192448746468, + "grad_norm": 0.9262595570800953, + "learning_rate": 5.0640632842260835e-08, + "loss": 0.8495, + "step": 7737 + }, + { + "epoch": 0.9304394877652858, + "grad_norm": 1.510366697690637, + "learning_rate": 5.0466598195009426e-08, + "loss": 0.9639, + "step": 7738 + }, + { + "epoch": 0.930559730655925, + "grad_norm": 2.2365058395203756, + "learning_rate": 5.0292859290425036e-08, + "loss": 0.9404, + "step": 7739 + }, + { + "epoch": 0.9306799735465641, + "grad_norm": 1.9148047428834984, + "learning_rate": 5.011941615486348e-08, + "loss": 1.0159, + "step": 7740 + }, + { + "epoch": 0.9308002164372031, + "grad_norm": 2.1310270892050003, + "learning_rate": 4.994626881463659e-08, + "loss": 1.0785, + "step": 7741 + }, + { + "epoch": 0.9309204593278423, + "grad_norm": 1.813219388065154, + "learning_rate": 4.9773417296009814e-08, + "loss": 0.9458, + "step": 7742 + }, + { + "epoch": 0.9310407022184813, + "grad_norm": 1.834786366700176, + "learning_rate": 4.960086162520527e-08, + "loss": 0.8862, + "step": 7743 + }, + { + "epoch": 0.9311609451091204, + "grad_norm": 2.0668317216160865, + "learning_rate": 4.942860182839936e-08, + "loss": 1.061, + "step": 7744 + }, + { + "epoch": 0.9312811879997596, + "grad_norm": 2.1951224356808425, + "learning_rate": 4.925663793172341e-08, + "loss": 1.0326, + "step": 7745 + }, + { + "epoch": 0.9314014308903986, + "grad_norm": 0.7954817915869979, + "learning_rate": 4.908496996126477e-08, + "loss": 0.8026, + "step": 7746 + }, + { + "epoch": 0.9315216737810377, + "grad_norm": 1.9141155002029175, + "learning_rate": 4.89135979430646e-08, + "loss": 1.0006, + "step": 7747 + }, + { + "epoch": 0.9316419166716768, + "grad_norm": 1.9050280297087048, + "learning_rate": 4.874252190312078e-08, + "loss": 1.0953, + "step": 7748 + }, + { + "epoch": 0.9317621595623159, + "grad_norm": 1.5710008924786456, + "learning_rate": 4.857174186738477e-08, + "loss": 0.884, + "step": 7749 + }, + { + "epoch": 0.931882402452955, + "grad_norm": 3.237872016182275, + "learning_rate": 4.840125786176408e-08, + "loss": 0.9696, + "step": 7750 + }, + { + "epoch": 0.932002645343594, + "grad_norm": 1.8593531374888221, + "learning_rate": 4.823106991212067e-08, + "loss": 1.0087, + "step": 7751 + }, + { + "epoch": 0.9321228882342332, + "grad_norm": 2.174609432011462, + "learning_rate": 4.806117804427212e-08, + "loss": 1.069, + "step": 7752 + }, + { + "epoch": 0.9322431311248722, + "grad_norm": 2.0661779035946974, + "learning_rate": 4.7891582283990926e-08, + "loss": 0.8784, + "step": 7753 + }, + { + "epoch": 0.9323633740155113, + "grad_norm": 1.8111055205687274, + "learning_rate": 4.772228265700473e-08, + "loss": 0.9643, + "step": 7754 + }, + { + "epoch": 0.9324836169061504, + "grad_norm": 2.4777095797646895, + "learning_rate": 4.75532791889961e-08, + "loss": 0.9966, + "step": 7755 + }, + { + "epoch": 0.9326038597967895, + "grad_norm": 1.9310589445847937, + "learning_rate": 4.738457190560252e-08, + "loss": 0.8945, + "step": 7756 + }, + { + "epoch": 0.9327241026874286, + "grad_norm": 2.63605189049779, + "learning_rate": 4.721616083241664e-08, + "loss": 1.0272, + "step": 7757 + }, + { + "epoch": 0.9328443455780677, + "grad_norm": 1.684889858732484, + "learning_rate": 4.7048045994986684e-08, + "loss": 1.0145, + "step": 7758 + }, + { + "epoch": 0.9329645884687068, + "grad_norm": 1.9139431663214796, + "learning_rate": 4.688022741881559e-08, + "loss": 1.1468, + "step": 7759 + }, + { + "epoch": 0.9330848313593458, + "grad_norm": 7.4943628106723175, + "learning_rate": 4.671270512936076e-08, + "loss": 0.9893, + "step": 7760 + }, + { + "epoch": 0.933205074249985, + "grad_norm": 1.7052132639253423, + "learning_rate": 4.6545479152035884e-08, + "loss": 1.0686, + "step": 7761 + }, + { + "epoch": 0.9333253171406241, + "grad_norm": 1.958987539332564, + "learning_rate": 4.637854951220821e-08, + "loss": 0.9962, + "step": 7762 + }, + { + "epoch": 0.9334455600312631, + "grad_norm": 1.9955340830722188, + "learning_rate": 4.621191623520171e-08, + "loss": 0.9848, + "step": 7763 + }, + { + "epoch": 0.9335658029219023, + "grad_norm": 2.3067932853602326, + "learning_rate": 4.604557934629372e-08, + "loss": 1.0833, + "step": 7764 + }, + { + "epoch": 0.9336860458125413, + "grad_norm": 1.7443024665953732, + "learning_rate": 4.587953887071805e-08, + "loss": 1.0425, + "step": 7765 + }, + { + "epoch": 0.9338062887031804, + "grad_norm": 1.7804452517115041, + "learning_rate": 4.5713794833662554e-08, + "loss": 1.0943, + "step": 7766 + }, + { + "epoch": 0.9339265315938196, + "grad_norm": 2.038949079026446, + "learning_rate": 4.5548347260270236e-08, + "loss": 0.8677, + "step": 7767 + }, + { + "epoch": 0.9340467744844586, + "grad_norm": 1.6822345672774834, + "learning_rate": 4.538319617564012e-08, + "loss": 0.9333, + "step": 7768 + }, + { + "epoch": 0.9341670173750977, + "grad_norm": 2.009368531022404, + "learning_rate": 4.521834160482485e-08, + "loss": 0.9805, + "step": 7769 + }, + { + "epoch": 0.9342872602657368, + "grad_norm": 1.7135456707413879, + "learning_rate": 4.5053783572832846e-08, + "loss": 1.0594, + "step": 7770 + }, + { + "epoch": 0.9344075031563759, + "grad_norm": 1.6641566290223975, + "learning_rate": 4.488952210462771e-08, + "loss": 0.998, + "step": 7771 + }, + { + "epoch": 0.9345277460470149, + "grad_norm": 1.925522111708781, + "learning_rate": 4.4725557225127495e-08, + "loss": 1.0953, + "step": 7772 + }, + { + "epoch": 0.9346479889376541, + "grad_norm": 1.5644387233799908, + "learning_rate": 4.456188895920565e-08, + "loss": 1.0283, + "step": 7773 + }, + { + "epoch": 0.9347682318282932, + "grad_norm": 2.183313996712279, + "learning_rate": 4.439851733169031e-08, + "loss": 1.0899, + "step": 7774 + }, + { + "epoch": 0.9348884747189322, + "grad_norm": 2.2680959972672725, + "learning_rate": 4.4235442367365204e-08, + "loss": 0.9317, + "step": 7775 + }, + { + "epoch": 0.9350087176095714, + "grad_norm": 2.0674174093704134, + "learning_rate": 4.4072664090968545e-08, + "loss": 1.03, + "step": 7776 + }, + { + "epoch": 0.9351289605002104, + "grad_norm": 2.232567009000293, + "learning_rate": 4.391018252719347e-08, + "loss": 1.085, + "step": 7777 + }, + { + "epoch": 0.9352492033908495, + "grad_norm": 1.858922053713449, + "learning_rate": 4.374799770068849e-08, + "loss": 0.9297, + "step": 7778 + }, + { + "epoch": 0.9353694462814887, + "grad_norm": 2.1700017521231474, + "learning_rate": 4.358610963605658e-08, + "loss": 0.9836, + "step": 7779 + }, + { + "epoch": 0.9354896891721277, + "grad_norm": 2.1877301776220315, + "learning_rate": 4.342451835785677e-08, + "loss": 0.9231, + "step": 7780 + }, + { + "epoch": 0.9356099320627668, + "grad_norm": 1.6076117921344055, + "learning_rate": 4.3263223890601665e-08, + "loss": 0.989, + "step": 7781 + }, + { + "epoch": 0.9357301749534058, + "grad_norm": 1.903512396008285, + "learning_rate": 4.31022262587597e-08, + "loss": 1.0286, + "step": 7782 + }, + { + "epoch": 0.935850417844045, + "grad_norm": 1.587023433731029, + "learning_rate": 4.2941525486754225e-08, + "loss": 0.8993, + "step": 7783 + }, + { + "epoch": 0.935970660734684, + "grad_norm": 2.174166342034784, + "learning_rate": 4.278112159896286e-08, + "loss": 1.0306, + "step": 7784 + }, + { + "epoch": 0.9360909036253231, + "grad_norm": 1.8576072816942606, + "learning_rate": 4.2621014619719896e-08, + "loss": 0.913, + "step": 7785 + }, + { + "epoch": 0.9362111465159623, + "grad_norm": 0.7277246840072837, + "learning_rate": 4.246120457331215e-08, + "loss": 0.8233, + "step": 7786 + }, + { + "epoch": 0.9363313894066013, + "grad_norm": 2.316267604441888, + "learning_rate": 4.2301691483983325e-08, + "loss": 0.9585, + "step": 7787 + }, + { + "epoch": 0.9364516322972404, + "grad_norm": 2.503018273829813, + "learning_rate": 4.214247537593163e-08, + "loss": 0.9985, + "step": 7788 + }, + { + "epoch": 0.9365718751878795, + "grad_norm": 1.9217549108187182, + "learning_rate": 4.1983556273309293e-08, + "loss": 1.0404, + "step": 7789 + }, + { + "epoch": 0.9366921180785186, + "grad_norm": 2.3979794937860515, + "learning_rate": 4.182493420022526e-08, + "loss": 0.9228, + "step": 7790 + }, + { + "epoch": 0.9368123609691577, + "grad_norm": 1.9358171106918554, + "learning_rate": 4.166660918074139e-08, + "loss": 1.0233, + "step": 7791 + }, + { + "epoch": 0.9369326038597968, + "grad_norm": 1.4819405117219087, + "learning_rate": 4.15085812388758e-08, + "loss": 0.9733, + "step": 7792 + }, + { + "epoch": 0.9370528467504359, + "grad_norm": 1.8901768302948074, + "learning_rate": 4.135085039860153e-08, + "loss": 1.0226, + "step": 7793 + }, + { + "epoch": 0.9371730896410749, + "grad_norm": 2.313011937567041, + "learning_rate": 4.1193416683845906e-08, + "loss": 1.0235, + "step": 7794 + }, + { + "epoch": 0.9372933325317141, + "grad_norm": 2.2134164453130096, + "learning_rate": 4.103628011849136e-08, + "loss": 1.0734, + "step": 7795 + }, + { + "epoch": 0.9374135754223532, + "grad_norm": 1.8916942000660997, + "learning_rate": 4.0879440726375506e-08, + "loss": 0.9948, + "step": 7796 + }, + { + "epoch": 0.9375338183129922, + "grad_norm": 2.6290580626322986, + "learning_rate": 4.0722898531291074e-08, + "loss": 0.7969, + "step": 7797 + }, + { + "epoch": 0.9376540612036314, + "grad_norm": 3.3302793022535013, + "learning_rate": 4.0566653556985295e-08, + "loss": 1.0042, + "step": 7798 + }, + { + "epoch": 0.9377743040942704, + "grad_norm": 2.392254641716259, + "learning_rate": 4.0410705827159886e-08, + "loss": 1.0523, + "step": 7799 + }, + { + "epoch": 0.9378945469849095, + "grad_norm": 2.643990984094804, + "learning_rate": 4.0255055365472356e-08, + "loss": 0.9461, + "step": 7800 + }, + { + "epoch": 0.9380147898755486, + "grad_norm": 2.2070833183155583, + "learning_rate": 4.009970219553471e-08, + "loss": 0.9841, + "step": 7801 + }, + { + "epoch": 0.9381350327661877, + "grad_norm": 2.5207055955369597, + "learning_rate": 3.99446463409141e-08, + "loss": 1.0077, + "step": 7802 + }, + { + "epoch": 0.9382552756568268, + "grad_norm": 2.567566120369776, + "learning_rate": 3.978988782513215e-08, + "loss": 0.9329, + "step": 7803 + }, + { + "epoch": 0.9383755185474659, + "grad_norm": 1.7193480404869084, + "learning_rate": 3.963542667166586e-08, + "loss": 1.0015, + "step": 7804 + }, + { + "epoch": 0.938495761438105, + "grad_norm": 1.9330714897990897, + "learning_rate": 3.9481262903946486e-08, + "loss": 0.9186, + "step": 7805 + }, + { + "epoch": 0.938616004328744, + "grad_norm": 0.7910985832511243, + "learning_rate": 3.932739654536066e-08, + "loss": 0.783, + "step": 7806 + }, + { + "epoch": 0.9387362472193832, + "grad_norm": 2.7839744168399583, + "learning_rate": 3.917382761925014e-08, + "loss": 0.9762, + "step": 7807 + }, + { + "epoch": 0.9388564901100223, + "grad_norm": 1.7365810351698034, + "learning_rate": 3.9020556148910754e-08, + "loss": 1.0301, + "step": 7808 + }, + { + "epoch": 0.9389767330006613, + "grad_norm": 0.7242591607959942, + "learning_rate": 3.8867582157593895e-08, + "loss": 0.8046, + "step": 7809 + }, + { + "epoch": 0.9390969758913005, + "grad_norm": 1.678452148373579, + "learning_rate": 3.871490566850544e-08, + "loss": 1.0021, + "step": 7810 + }, + { + "epoch": 0.9392172187819395, + "grad_norm": 1.7782389974673392, + "learning_rate": 3.856252670480642e-08, + "loss": 0.942, + "step": 7811 + }, + { + "epoch": 0.9393374616725786, + "grad_norm": 1.782409023194271, + "learning_rate": 3.841044528961279e-08, + "loss": 1.0514, + "step": 7812 + }, + { + "epoch": 0.9394577045632178, + "grad_norm": 2.6009971760489194, + "learning_rate": 3.825866144599477e-08, + "loss": 1.0258, + "step": 7813 + }, + { + "epoch": 0.9395779474538568, + "grad_norm": 3.121076791500001, + "learning_rate": 3.8107175196978145e-08, + "loss": 0.9944, + "step": 7814 + }, + { + "epoch": 0.9396981903444959, + "grad_norm": 2.938382442307043, + "learning_rate": 3.7955986565542996e-08, + "loss": 1.0022, + "step": 7815 + }, + { + "epoch": 0.9398184332351349, + "grad_norm": 1.8635783408493751, + "learning_rate": 3.780509557462497e-08, + "loss": 0.9157, + "step": 7816 + }, + { + "epoch": 0.9399386761257741, + "grad_norm": 2.4159580923892094, + "learning_rate": 3.765450224711375e-08, + "loss": 0.9935, + "step": 7817 + }, + { + "epoch": 0.9400589190164131, + "grad_norm": 1.6696645795022524, + "learning_rate": 3.750420660585396e-08, + "loss": 1.035, + "step": 7818 + }, + { + "epoch": 0.9401791619070522, + "grad_norm": 1.8882126984721903, + "learning_rate": 3.735420867364603e-08, + "loss": 1.0315, + "step": 7819 + }, + { + "epoch": 0.9402994047976914, + "grad_norm": 3.4203001256503685, + "learning_rate": 3.7204508473244186e-08, + "loss": 0.8579, + "step": 7820 + }, + { + "epoch": 0.9404196476883304, + "grad_norm": 2.004251967086674, + "learning_rate": 3.7055106027357395e-08, + "loss": 0.9306, + "step": 7821 + }, + { + "epoch": 0.9405398905789695, + "grad_norm": 2.1535556932718127, + "learning_rate": 3.690600135865063e-08, + "loss": 0.9596, + "step": 7822 + }, + { + "epoch": 0.9406601334696086, + "grad_norm": 0.7984463775200958, + "learning_rate": 3.675719448974246e-08, + "loss": 0.8181, + "step": 7823 + }, + { + "epoch": 0.9407803763602477, + "grad_norm": 2.134707772046022, + "learning_rate": 3.6608685443207054e-08, + "loss": 0.8366, + "step": 7824 + }, + { + "epoch": 0.9409006192508867, + "grad_norm": 2.7332040077721618, + "learning_rate": 3.646047424157306e-08, + "loss": 0.9081, + "step": 7825 + }, + { + "epoch": 0.9410208621415259, + "grad_norm": 3.2156561639608388, + "learning_rate": 3.631256090732382e-08, + "loss": 0.9232, + "step": 7826 + }, + { + "epoch": 0.941141105032165, + "grad_norm": 1.8329920089763931, + "learning_rate": 3.6164945462897833e-08, + "loss": 1.0611, + "step": 7827 + }, + { + "epoch": 0.941261347922804, + "grad_norm": 1.8709893247258027, + "learning_rate": 3.6017627930687856e-08, + "loss": 0.9921, + "step": 7828 + }, + { + "epoch": 0.9413815908134432, + "grad_norm": 2.6253599192602066, + "learning_rate": 3.587060833304267e-08, + "loss": 1.0106, + "step": 7829 + }, + { + "epoch": 0.9415018337040822, + "grad_norm": 3.324949370264055, + "learning_rate": 3.5723886692264225e-08, + "loss": 0.8773, + "step": 7830 + }, + { + "epoch": 0.9416220765947213, + "grad_norm": 2.2384863954172833, + "learning_rate": 3.557746303061071e-08, + "loss": 0.8532, + "step": 7831 + }, + { + "epoch": 0.9417423194853605, + "grad_norm": 2.4457207969583945, + "learning_rate": 3.543133737029391e-08, + "loss": 0.9593, + "step": 7832 + }, + { + "epoch": 0.9418625623759995, + "grad_norm": 1.8231173668231786, + "learning_rate": 3.5285509733481214e-08, + "loss": 0.9244, + "step": 7833 + }, + { + "epoch": 0.9419828052666386, + "grad_norm": 1.7454395818249537, + "learning_rate": 3.513998014229469e-08, + "loss": 1.0028, + "step": 7834 + }, + { + "epoch": 0.9421030481572777, + "grad_norm": 3.1069780628619617, + "learning_rate": 3.499474861881069e-08, + "loss": 1.1008, + "step": 7835 + }, + { + "epoch": 0.9422232910479168, + "grad_norm": 2.107009578674009, + "learning_rate": 3.4849815185061136e-08, + "loss": 0.9186, + "step": 7836 + }, + { + "epoch": 0.9423435339385559, + "grad_norm": 2.220898944202849, + "learning_rate": 3.470517986303223e-08, + "loss": 0.9943, + "step": 7837 + }, + { + "epoch": 0.942463776829195, + "grad_norm": 1.7200695867656168, + "learning_rate": 3.4560842674664856e-08, + "loss": 1.0295, + "step": 7838 + }, + { + "epoch": 0.9425840197198341, + "grad_norm": 1.859092947460006, + "learning_rate": 3.441680364185506e-08, + "loss": 0.9893, + "step": 7839 + }, + { + "epoch": 0.9427042626104731, + "grad_norm": 2.7266635821386425, + "learning_rate": 3.427306278645314e-08, + "loss": 0.9834, + "step": 7840 + }, + { + "epoch": 0.9428245055011123, + "grad_norm": 1.849415431584633, + "learning_rate": 3.4129620130264767e-08, + "loss": 0.9678, + "step": 7841 + }, + { + "epoch": 0.9429447483917514, + "grad_norm": 2.2178191260499136, + "learning_rate": 3.398647569505009e-08, + "loss": 1.0155, + "step": 7842 + }, + { + "epoch": 0.9430649912823904, + "grad_norm": 2.4269557695606165, + "learning_rate": 3.384362950252373e-08, + "loss": 0.9781, + "step": 7843 + }, + { + "epoch": 0.9431852341730296, + "grad_norm": 2.7155751073400154, + "learning_rate": 3.3701081574355473e-08, + "loss": 0.8066, + "step": 7844 + }, + { + "epoch": 0.9433054770636686, + "grad_norm": 0.6481529668661655, + "learning_rate": 3.3558831932169796e-08, + "loss": 0.7541, + "step": 7845 + }, + { + "epoch": 0.9434257199543077, + "grad_norm": 1.9945198916430464, + "learning_rate": 3.341688059754588e-08, + "loss": 1.124, + "step": 7846 + }, + { + "epoch": 0.9435459628449467, + "grad_norm": 2.1731934422051924, + "learning_rate": 3.327522759201762e-08, + "loss": 1.0152, + "step": 7847 + }, + { + "epoch": 0.9436662057355859, + "grad_norm": 2.314479290134273, + "learning_rate": 3.313387293707359e-08, + "loss": 0.8989, + "step": 7848 + }, + { + "epoch": 0.943786448626225, + "grad_norm": 2.222829051080161, + "learning_rate": 3.29928166541571e-08, + "loss": 0.9197, + "step": 7849 + }, + { + "epoch": 0.943906691516864, + "grad_norm": 2.108180746454843, + "learning_rate": 3.2852058764666346e-08, + "loss": 1.0405, + "step": 7850 + }, + { + "epoch": 0.9440269344075032, + "grad_norm": 1.826883465032447, + "learning_rate": 3.2711599289954264e-08, + "loss": 0.9223, + "step": 7851 + }, + { + "epoch": 0.9441471772981422, + "grad_norm": 1.9222186021764598, + "learning_rate": 3.257143825132847e-08, + "loss": 1.013, + "step": 7852 + }, + { + "epoch": 0.9442674201887813, + "grad_norm": 1.707025593414362, + "learning_rate": 3.243157567005106e-08, + "loss": 0.9981, + "step": 7853 + }, + { + "epoch": 0.9443876630794205, + "grad_norm": 2.185560292875199, + "learning_rate": 3.2292011567339296e-08, + "loss": 0.874, + "step": 7854 + }, + { + "epoch": 0.9445079059700595, + "grad_norm": 2.2262669417321925, + "learning_rate": 3.21527459643649e-08, + "loss": 0.7942, + "step": 7855 + }, + { + "epoch": 0.9446281488606986, + "grad_norm": 1.879644998147597, + "learning_rate": 3.2013778882254536e-08, + "loss": 0.9808, + "step": 7856 + }, + { + "epoch": 0.9447483917513377, + "grad_norm": 1.8772401143716873, + "learning_rate": 3.1875110342088676e-08, + "loss": 0.9933, + "step": 7857 + }, + { + "epoch": 0.9448686346419768, + "grad_norm": 1.786665264041312, + "learning_rate": 3.1736740364904035e-08, + "loss": 0.8894, + "step": 7858 + }, + { + "epoch": 0.9449888775326158, + "grad_norm": 2.4018634595154484, + "learning_rate": 3.159866897169094e-08, + "loss": 1.0061, + "step": 7859 + }, + { + "epoch": 0.945109120423255, + "grad_norm": 2.521804328315839, + "learning_rate": 3.146089618339487e-08, + "loss": 0.9921, + "step": 7860 + }, + { + "epoch": 0.9452293633138941, + "grad_norm": 7.286699474199201, + "learning_rate": 3.132342202091554e-08, + "loss": 0.9129, + "step": 7861 + }, + { + "epoch": 0.9453496062045331, + "grad_norm": 2.275715633642006, + "learning_rate": 3.1186246505107595e-08, + "loss": 0.921, + "step": 7862 + }, + { + "epoch": 0.9454698490951723, + "grad_norm": 1.7555272824724215, + "learning_rate": 3.104936965678084e-08, + "loss": 1.0729, + "step": 7863 + }, + { + "epoch": 0.9455900919858113, + "grad_norm": 2.088072409357192, + "learning_rate": 3.091279149669956e-08, + "loss": 1.0561, + "step": 7864 + }, + { + "epoch": 0.9457103348764504, + "grad_norm": 1.8159161306373461, + "learning_rate": 3.0776512045581624e-08, + "loss": 0.9754, + "step": 7865 + }, + { + "epoch": 0.9458305777670896, + "grad_norm": 2.492304114600002, + "learning_rate": 3.0640531324101384e-08, + "loss": 1.0119, + "step": 7866 + }, + { + "epoch": 0.9459508206577286, + "grad_norm": 1.907801950036574, + "learning_rate": 3.0504849352886554e-08, + "loss": 0.999, + "step": 7867 + }, + { + "epoch": 0.9460710635483677, + "grad_norm": 3.3356630449160702, + "learning_rate": 3.036946615252023e-08, + "loss": 0.9534, + "step": 7868 + }, + { + "epoch": 0.9461913064390068, + "grad_norm": 2.2331902624203215, + "learning_rate": 3.0234381743539984e-08, + "loss": 0.911, + "step": 7869 + }, + { + "epoch": 0.9463115493296459, + "grad_norm": 2.4776282679730293, + "learning_rate": 3.0099596146437863e-08, + "loss": 1.0357, + "step": 7870 + }, + { + "epoch": 0.946431792220285, + "grad_norm": 0.7769262448240456, + "learning_rate": 2.996510938166086e-08, + "loss": 0.8376, + "step": 7871 + }, + { + "epoch": 0.9465520351109241, + "grad_norm": 1.8681125025556116, + "learning_rate": 2.983092146960997e-08, + "loss": 0.9749, + "step": 7872 + }, + { + "epoch": 0.9466722780015632, + "grad_norm": 2.221332605100026, + "learning_rate": 2.9697032430642256e-08, + "loss": 1.0389, + "step": 7873 + }, + { + "epoch": 0.9467925208922022, + "grad_norm": 2.337246021201672, + "learning_rate": 2.9563442285067906e-08, + "loss": 0.9786, + "step": 7874 + }, + { + "epoch": 0.9469127637828414, + "grad_norm": 2.2253069633733213, + "learning_rate": 2.943015105315294e-08, + "loss": 1.0324, + "step": 7875 + }, + { + "epoch": 0.9470330066734804, + "grad_norm": 2.9481584070159608, + "learning_rate": 2.929715875511718e-08, + "loss": 0.9022, + "step": 7876 + }, + { + "epoch": 0.9471532495641195, + "grad_norm": 1.9247181029564793, + "learning_rate": 2.9164465411135375e-08, + "loss": 0.9359, + "step": 7877 + }, + { + "epoch": 0.9472734924547586, + "grad_norm": 1.9386109889237015, + "learning_rate": 2.9032071041337426e-08, + "loss": 1.0473, + "step": 7878 + }, + { + "epoch": 0.9473937353453977, + "grad_norm": 1.7396156932386972, + "learning_rate": 2.889997566580704e-08, + "loss": 0.9651, + "step": 7879 + }, + { + "epoch": 0.9475139782360368, + "grad_norm": 1.708907224821071, + "learning_rate": 2.8768179304583086e-08, + "loss": 0.9385, + "step": 7880 + }, + { + "epoch": 0.9476342211266758, + "grad_norm": 1.7584196949340918, + "learning_rate": 2.8636681977659117e-08, + "loss": 0.9716, + "step": 7881 + }, + { + "epoch": 0.947754464017315, + "grad_norm": 1.9550855638360602, + "learning_rate": 2.850548370498318e-08, + "loss": 1.0131, + "step": 7882 + }, + { + "epoch": 0.9478747069079541, + "grad_norm": 2.0066407343096015, + "learning_rate": 2.8374584506457798e-08, + "loss": 0.951, + "step": 7883 + }, + { + "epoch": 0.9479949497985931, + "grad_norm": 2.297039830528027, + "learning_rate": 2.824398440193998e-08, + "loss": 0.911, + "step": 7884 + }, + { + "epoch": 0.9481151926892323, + "grad_norm": 2.070481088359867, + "learning_rate": 2.811368341124232e-08, + "loss": 0.9523, + "step": 7885 + }, + { + "epoch": 0.9482354355798713, + "grad_norm": 2.2139778934053957, + "learning_rate": 2.7983681554131222e-08, + "loss": 0.9183, + "step": 7886 + }, + { + "epoch": 0.9483556784705104, + "grad_norm": 2.225116460945473, + "learning_rate": 2.7853978850327365e-08, + "loss": 0.9388, + "step": 7887 + }, + { + "epoch": 0.9484759213611496, + "grad_norm": 1.7628457805627733, + "learning_rate": 2.7724575319507225e-08, + "loss": 1.1078, + "step": 7888 + }, + { + "epoch": 0.9485961642517886, + "grad_norm": 1.8879876776228475, + "learning_rate": 2.759547098130044e-08, + "loss": 1.0085, + "step": 7889 + }, + { + "epoch": 0.9487164071424277, + "grad_norm": 1.7921560269745942, + "learning_rate": 2.746666585529267e-08, + "loss": 1.0057, + "step": 7890 + }, + { + "epoch": 0.9488366500330668, + "grad_norm": 2.496250965498354, + "learning_rate": 2.73381599610234e-08, + "loss": 0.9798, + "step": 7891 + }, + { + "epoch": 0.9489568929237059, + "grad_norm": 2.493002635662156, + "learning_rate": 2.7209953317987033e-08, + "loss": 0.9526, + "step": 7892 + }, + { + "epoch": 0.9490771358143449, + "grad_norm": 1.9715818595328756, + "learning_rate": 2.7082045945631793e-08, + "loss": 1.02, + "step": 7893 + }, + { + "epoch": 0.9491973787049841, + "grad_norm": 2.523301348246237, + "learning_rate": 2.6954437863361712e-08, + "loss": 0.9317, + "step": 7894 + }, + { + "epoch": 0.9493176215956232, + "grad_norm": 1.9837126339792728, + "learning_rate": 2.6827129090534862e-08, + "loss": 0.9443, + "step": 7895 + }, + { + "epoch": 0.9494378644862622, + "grad_norm": 1.983349212080706, + "learning_rate": 2.670011964646335e-08, + "loss": 1.0165, + "step": 7896 + }, + { + "epoch": 0.9495581073769014, + "grad_norm": 2.480315724856556, + "learning_rate": 2.657340955041487e-08, + "loss": 0.9165, + "step": 7897 + }, + { + "epoch": 0.9496783502675404, + "grad_norm": 2.1012702369963554, + "learning_rate": 2.6446998821611167e-08, + "loss": 0.9474, + "step": 7898 + }, + { + "epoch": 0.9497985931581795, + "grad_norm": 2.5623669473484414, + "learning_rate": 2.6320887479228228e-08, + "loss": 0.9545, + "step": 7899 + }, + { + "epoch": 0.9499188360488187, + "grad_norm": 2.406892864009443, + "learning_rate": 2.619507554239786e-08, + "loss": 0.9627, + "step": 7900 + }, + { + "epoch": 0.9500390789394577, + "grad_norm": 1.687155582460278, + "learning_rate": 2.606956303020502e-08, + "loss": 0.9348, + "step": 7901 + }, + { + "epoch": 0.9501593218300968, + "grad_norm": 2.1734912813554543, + "learning_rate": 2.5944349961690036e-08, + "loss": 1.0783, + "step": 7902 + }, + { + "epoch": 0.9502795647207359, + "grad_norm": 1.9387830362266063, + "learning_rate": 2.581943635584749e-08, + "loss": 0.9694, + "step": 7903 + }, + { + "epoch": 0.950399807611375, + "grad_norm": 1.5528828483947599, + "learning_rate": 2.569482223162689e-08, + "loss": 0.891, + "step": 7904 + }, + { + "epoch": 0.950520050502014, + "grad_norm": 1.7321659488290482, + "learning_rate": 2.5570507607932e-08, + "loss": 0.9632, + "step": 7905 + }, + { + "epoch": 0.9506402933926532, + "grad_norm": 2.4603672880751413, + "learning_rate": 2.54464925036213e-08, + "loss": 0.8691, + "step": 7906 + }, + { + "epoch": 0.9507605362832923, + "grad_norm": 2.081975138124623, + "learning_rate": 2.532277693750773e-08, + "loss": 0.8448, + "step": 7907 + }, + { + "epoch": 0.9508807791739313, + "grad_norm": 5.010136057820295, + "learning_rate": 2.5199360928358948e-08, + "loss": 0.9951, + "step": 7908 + }, + { + "epoch": 0.9510010220645704, + "grad_norm": 1.702830437901596, + "learning_rate": 2.507624449489665e-08, + "loss": 1.1083, + "step": 7909 + }, + { + "epoch": 0.9511212649552095, + "grad_norm": 1.849332214665, + "learning_rate": 2.495342765579811e-08, + "loss": 0.8829, + "step": 7910 + }, + { + "epoch": 0.9512415078458486, + "grad_norm": 1.7809118652640783, + "learning_rate": 2.4830910429693984e-08, + "loss": 0.9474, + "step": 7911 + }, + { + "epoch": 0.9513617507364877, + "grad_norm": 2.0082404129671363, + "learning_rate": 2.470869283517052e-08, + "loss": 1.0322, + "step": 7912 + }, + { + "epoch": 0.9514819936271268, + "grad_norm": 1.9098501135648236, + "learning_rate": 2.458677489076777e-08, + "loss": 1.0084, + "step": 7913 + }, + { + "epoch": 0.9516022365177659, + "grad_norm": 1.668532836728121, + "learning_rate": 2.446515661498072e-08, + "loss": 1.0651, + "step": 7914 + }, + { + "epoch": 0.9517224794084049, + "grad_norm": 2.5896996735647697, + "learning_rate": 2.434383802625861e-08, + "loss": 0.9806, + "step": 7915 + }, + { + "epoch": 0.9518427222990441, + "grad_norm": 2.408821218454285, + "learning_rate": 2.4222819143005168e-08, + "loss": 0.9732, + "step": 7916 + }, + { + "epoch": 0.9519629651896832, + "grad_norm": 1.8399405484339, + "learning_rate": 2.4102099983579706e-08, + "loss": 1.0476, + "step": 7917 + }, + { + "epoch": 0.9520832080803222, + "grad_norm": 2.014597786263422, + "learning_rate": 2.3981680566294236e-08, + "loss": 1.0089, + "step": 7918 + }, + { + "epoch": 0.9522034509709614, + "grad_norm": 1.827038499143602, + "learning_rate": 2.3861560909416822e-08, + "loss": 0.971, + "step": 7919 + }, + { + "epoch": 0.9523236938616004, + "grad_norm": 2.069754356146545, + "learning_rate": 2.3741741031169325e-08, + "loss": 1.0627, + "step": 7920 + }, + { + "epoch": 0.9524439367522395, + "grad_norm": 2.611085981784679, + "learning_rate": 2.3622220949728544e-08, + "loss": 0.9474, + "step": 7921 + }, + { + "epoch": 0.9525641796428787, + "grad_norm": 2.3231524930799345, + "learning_rate": 2.3503000683225526e-08, + "loss": 0.8489, + "step": 7922 + }, + { + "epoch": 0.9526844225335177, + "grad_norm": 2.4217761604075148, + "learning_rate": 2.3384080249745585e-08, + "loss": 1.0763, + "step": 7923 + }, + { + "epoch": 0.9528046654241568, + "grad_norm": 3.4258722795595418, + "learning_rate": 2.3265459667329178e-08, + "loss": 1.0645, + "step": 7924 + }, + { + "epoch": 0.9529249083147959, + "grad_norm": 2.903643187563023, + "learning_rate": 2.31471389539708e-08, + "loss": 1.1042, + "step": 7925 + }, + { + "epoch": 0.953045151205435, + "grad_norm": 2.0761789947990574, + "learning_rate": 2.3029118127619872e-08, + "loss": 0.968, + "step": 7926 + }, + { + "epoch": 0.953165394096074, + "grad_norm": 2.2316367703696143, + "learning_rate": 2.2911397206179628e-08, + "loss": 1.103, + "step": 7927 + }, + { + "epoch": 0.9532856369867132, + "grad_norm": 1.7883675903968843, + "learning_rate": 2.279397620750845e-08, + "loss": 0.8636, + "step": 7928 + }, + { + "epoch": 0.9534058798773523, + "grad_norm": 2.046239739766902, + "learning_rate": 2.2676855149419195e-08, + "loss": 1.024, + "step": 7929 + }, + { + "epoch": 0.9535261227679913, + "grad_norm": 2.2963329895339704, + "learning_rate": 2.2560034049678988e-08, + "loss": 0.9951, + "step": 7930 + }, + { + "epoch": 0.9536463656586305, + "grad_norm": 2.0091777271592335, + "learning_rate": 2.2443512926008988e-08, + "loss": 0.9896, + "step": 7931 + }, + { + "epoch": 0.9537666085492695, + "grad_norm": 2.353514308058501, + "learning_rate": 2.2327291796085946e-08, + "loss": 0.9314, + "step": 7932 + }, + { + "epoch": 0.9538868514399086, + "grad_norm": 2.6721128837473858, + "learning_rate": 2.2211370677540197e-08, + "loss": 1.0138, + "step": 7933 + }, + { + "epoch": 0.9540070943305478, + "grad_norm": 3.46554486084889, + "learning_rate": 2.2095749587957012e-08, + "loss": 1.0178, + "step": 7934 + }, + { + "epoch": 0.9541273372211868, + "grad_norm": 2.0871718965447763, + "learning_rate": 2.1980428544876138e-08, + "loss": 0.9323, + "step": 7935 + }, + { + "epoch": 0.9542475801118259, + "grad_norm": 1.838947312629682, + "learning_rate": 2.1865407565791584e-08, + "loss": 0.9766, + "step": 7936 + }, + { + "epoch": 0.954367823002465, + "grad_norm": 2.1505703413562967, + "learning_rate": 2.175068666815183e-08, + "loss": 1.0137, + "step": 7937 + }, + { + "epoch": 0.9544880658931041, + "grad_norm": 2.773846491355457, + "learning_rate": 2.163626586935985e-08, + "loss": 1.027, + "step": 7938 + }, + { + "epoch": 0.9546083087837431, + "grad_norm": 1.853458383558622, + "learning_rate": 2.1522145186773755e-08, + "loss": 0.8676, + "step": 7939 + }, + { + "epoch": 0.9547285516743822, + "grad_norm": 1.725925932345908, + "learning_rate": 2.140832463770481e-08, + "loss": 1.091, + "step": 7940 + }, + { + "epoch": 0.9548487945650214, + "grad_norm": 2.1234357349649464, + "learning_rate": 2.129480423941987e-08, + "loss": 0.9962, + "step": 7941 + }, + { + "epoch": 0.9549690374556604, + "grad_norm": 1.637081643162987, + "learning_rate": 2.1181584009140052e-08, + "loss": 1.0406, + "step": 7942 + }, + { + "epoch": 0.9550892803462995, + "grad_norm": 2.1607884728712823, + "learning_rate": 2.10686639640405e-08, + "loss": 1.0724, + "step": 7943 + }, + { + "epoch": 0.9552095232369386, + "grad_norm": 1.9212539413699623, + "learning_rate": 2.0956044121251294e-08, + "loss": 1.0527, + "step": 7944 + }, + { + "epoch": 0.9553297661275777, + "grad_norm": 1.803455741603663, + "learning_rate": 2.084372449785654e-08, + "loss": 1.0478, + "step": 7945 + }, + { + "epoch": 0.9554500090182168, + "grad_norm": 1.8775417266353784, + "learning_rate": 2.0731705110895282e-08, + "loss": 0.9237, + "step": 7946 + }, + { + "epoch": 0.9555702519088559, + "grad_norm": 1.8089858880280734, + "learning_rate": 2.0619985977360587e-08, + "loss": 1.1046, + "step": 7947 + }, + { + "epoch": 0.955690494799495, + "grad_norm": 1.7739846482026602, + "learning_rate": 2.0508567114200237e-08, + "loss": 1.003, + "step": 7948 + }, + { + "epoch": 0.955810737690134, + "grad_norm": 3.181511583047273, + "learning_rate": 2.0397448538316485e-08, + "loss": 1.0204, + "step": 7949 + }, + { + "epoch": 0.9559309805807732, + "grad_norm": 2.2371077465426454, + "learning_rate": 2.028663026656563e-08, + "loss": 0.9028, + "step": 7950 + }, + { + "epoch": 0.9560512234714122, + "grad_norm": 2.0948662131115157, + "learning_rate": 2.0176112315758885e-08, + "loss": 0.953, + "step": 7951 + }, + { + "epoch": 0.9561714663620513, + "grad_norm": 3.2160038761001686, + "learning_rate": 2.0065894702661957e-08, + "loss": 0.9277, + "step": 7952 + }, + { + "epoch": 0.9562917092526905, + "grad_norm": 1.6802594297916642, + "learning_rate": 1.9955977443994577e-08, + "loss": 1.0145, + "step": 7953 + }, + { + "epoch": 0.9564119521433295, + "grad_norm": 2.377396027450599, + "learning_rate": 1.9846360556430965e-08, + "loss": 0.8565, + "step": 7954 + }, + { + "epoch": 0.9565321950339686, + "grad_norm": 2.395897529958253, + "learning_rate": 1.973704405660004e-08, + "loss": 0.8495, + "step": 7955 + }, + { + "epoch": 0.9566524379246077, + "grad_norm": 4.972258091139512, + "learning_rate": 1.9628027961085203e-08, + "loss": 1.017, + "step": 7956 + }, + { + "epoch": 0.9567726808152468, + "grad_norm": 2.810248087998829, + "learning_rate": 1.9519312286423894e-08, + "loss": 1.0785, + "step": 7957 + }, + { + "epoch": 0.9568929237058859, + "grad_norm": 2.3416634999680292, + "learning_rate": 1.9410897049108255e-08, + "loss": 1.0154, + "step": 7958 + }, + { + "epoch": 0.957013166596525, + "grad_norm": 2.009261858707797, + "learning_rate": 1.9302782265584905e-08, + "loss": 1.1484, + "step": 7959 + }, + { + "epoch": 0.9571334094871641, + "grad_norm": 2.1363807424823333, + "learning_rate": 1.9194967952254282e-08, + "loss": 1.1101, + "step": 7960 + }, + { + "epoch": 0.9572536523778031, + "grad_norm": 2.4123221732108053, + "learning_rate": 1.9087454125472635e-08, + "loss": 1.0477, + "step": 7961 + }, + { + "epoch": 0.9573738952684423, + "grad_norm": 2.3965232740259372, + "learning_rate": 1.8980240801548696e-08, + "loss": 1.0214, + "step": 7962 + }, + { + "epoch": 0.9574941381590814, + "grad_norm": 2.4620969916977478, + "learning_rate": 1.8873327996747458e-08, + "loss": 0.9795, + "step": 7963 + }, + { + "epoch": 0.9576143810497204, + "grad_norm": 2.058947190810366, + "learning_rate": 1.8766715727287053e-08, + "loss": 0.8955, + "step": 7964 + }, + { + "epoch": 0.9577346239403596, + "grad_norm": 1.6974909274009091, + "learning_rate": 1.8660404009340546e-08, + "loss": 1.0312, + "step": 7965 + }, + { + "epoch": 0.9578548668309986, + "grad_norm": 0.8807824686991766, + "learning_rate": 1.8554392859035485e-08, + "loss": 0.833, + "step": 7966 + }, + { + "epoch": 0.9579751097216377, + "grad_norm": 1.717022322259122, + "learning_rate": 1.8448682292453444e-08, + "loss": 1.0254, + "step": 7967 + }, + { + "epoch": 0.9580953526122769, + "grad_norm": 1.957164280190429, + "learning_rate": 1.8343272325631154e-08, + "loss": 0.8954, + "step": 7968 + }, + { + "epoch": 0.9582155955029159, + "grad_norm": 2.585478312587131, + "learning_rate": 1.8238162974558492e-08, + "loss": 1.0163, + "step": 7969 + }, + { + "epoch": 0.958335838393555, + "grad_norm": 2.1800719438283074, + "learning_rate": 1.8133354255181144e-08, + "loss": 0.9877, + "step": 7970 + }, + { + "epoch": 0.958456081284194, + "grad_norm": 1.8460875741513882, + "learning_rate": 1.802884618339795e-08, + "loss": 0.9786, + "step": 7971 + }, + { + "epoch": 0.9585763241748332, + "grad_norm": 2.3158890917218486, + "learning_rate": 1.7924638775062894e-08, + "loss": 1.0439, + "step": 7972 + }, + { + "epoch": 0.9586965670654722, + "grad_norm": 1.9273394302796012, + "learning_rate": 1.7820732045984444e-08, + "loss": 1.0559, + "step": 7973 + }, + { + "epoch": 0.9588168099561113, + "grad_norm": 1.939901524826812, + "learning_rate": 1.7717126011924655e-08, + "loss": 0.98, + "step": 7974 + }, + { + "epoch": 0.9589370528467505, + "grad_norm": 2.4672734154399802, + "learning_rate": 1.7613820688600957e-08, + "loss": 1.0041, + "step": 7975 + }, + { + "epoch": 0.9590572957373895, + "grad_norm": 1.7553204312925494, + "learning_rate": 1.7510816091684588e-08, + "loss": 1.0245, + "step": 7976 + }, + { + "epoch": 0.9591775386280286, + "grad_norm": 2.746409412978387, + "learning_rate": 1.740811223680083e-08, + "loss": 1.0254, + "step": 7977 + }, + { + "epoch": 0.9592977815186677, + "grad_norm": 2.6869185062567746, + "learning_rate": 1.7305709139530334e-08, + "loss": 0.9795, + "step": 7978 + }, + { + "epoch": 0.9594180244093068, + "grad_norm": 2.526707753457177, + "learning_rate": 1.7203606815407334e-08, + "loss": 0.9875, + "step": 7979 + }, + { + "epoch": 0.9595382672999458, + "grad_norm": 1.9091662003999317, + "learning_rate": 1.7101805279920557e-08, + "loss": 1.0293, + "step": 7980 + }, + { + "epoch": 0.959658510190585, + "grad_norm": 2.1784166575895716, + "learning_rate": 1.7000304548513643e-08, + "loss": 1.044, + "step": 7981 + }, + { + "epoch": 0.9597787530812241, + "grad_norm": 2.1748427279382456, + "learning_rate": 1.6899104636583394e-08, + "loss": 1.0644, + "step": 7982 + }, + { + "epoch": 0.9598989959718631, + "grad_norm": 0.7325615101268868, + "learning_rate": 1.6798205559482638e-08, + "loss": 0.8561, + "step": 7983 + }, + { + "epoch": 0.9600192388625023, + "grad_norm": 1.84676037737682, + "learning_rate": 1.669760733251713e-08, + "loss": 1.0021, + "step": 7984 + }, + { + "epoch": 0.9601394817531413, + "grad_norm": 2.016028430397655, + "learning_rate": 1.659730997094755e-08, + "loss": 1.0608, + "step": 7985 + }, + { + "epoch": 0.9602597246437804, + "grad_norm": 2.0666944450644498, + "learning_rate": 1.6497313489989283e-08, + "loss": 0.8585, + "step": 7986 + }, + { + "epoch": 0.9603799675344196, + "grad_norm": 2.4526539733024606, + "learning_rate": 1.639761790481131e-08, + "loss": 0.9349, + "step": 7987 + }, + { + "epoch": 0.9605002104250586, + "grad_norm": 2.0393632104353427, + "learning_rate": 1.6298223230537754e-08, + "loss": 1.0273, + "step": 7988 + }, + { + "epoch": 0.9606204533156977, + "grad_norm": 2.006713804069447, + "learning_rate": 1.619912948224611e-08, + "loss": 0.9325, + "step": 7989 + }, + { + "epoch": 0.9607406962063368, + "grad_norm": 2.5591624785372136, + "learning_rate": 1.6100336674969682e-08, + "loss": 0.8487, + "step": 7990 + }, + { + "epoch": 0.9608609390969759, + "grad_norm": 1.8607590597009807, + "learning_rate": 1.600184482369449e-08, + "loss": 1.0001, + "step": 7991 + }, + { + "epoch": 0.960981181987615, + "grad_norm": 2.7137133543403484, + "learning_rate": 1.5903653943362126e-08, + "loss": 1.1256, + "step": 7992 + }, + { + "epoch": 0.9611014248782541, + "grad_norm": 3.7585948359714445, + "learning_rate": 1.580576404886802e-08, + "loss": 1.0075, + "step": 7993 + }, + { + "epoch": 0.9612216677688932, + "grad_norm": 2.0062331582191386, + "learning_rate": 1.570817515506162e-08, + "loss": 1.0341, + "step": 7994 + }, + { + "epoch": 0.9613419106595322, + "grad_norm": 2.0107972043529627, + "learning_rate": 1.561088727674753e-08, + "loss": 1.0547, + "step": 7995 + }, + { + "epoch": 0.9614621535501714, + "grad_norm": 2.421142762762957, + "learning_rate": 1.551390042868417e-08, + "loss": 0.9465, + "step": 7996 + }, + { + "epoch": 0.9615823964408104, + "grad_norm": 1.7592537629820573, + "learning_rate": 1.5417214625584207e-08, + "loss": 0.9431, + "step": 7997 + }, + { + "epoch": 0.9617026393314495, + "grad_norm": 1.7009430653699873, + "learning_rate": 1.5320829882114806e-08, + "loss": 1.0863, + "step": 7998 + }, + { + "epoch": 0.9618228822220887, + "grad_norm": 1.9694592848222972, + "learning_rate": 1.5224746212897378e-08, + "loss": 1.0236, + "step": 7999 + }, + { + "epoch": 0.9619431251127277, + "grad_norm": 1.6571577931186714, + "learning_rate": 1.512896363250804e-08, + "loss": 1.0093, + "step": 8000 + }, + { + "epoch": 0.9620633680033668, + "grad_norm": 1.8868480187907037, + "learning_rate": 1.503348215547673e-08, + "loss": 0.9929, + "step": 8001 + }, + { + "epoch": 0.962183610894006, + "grad_norm": 1.935444880164577, + "learning_rate": 1.4938301796288078e-08, + "loss": 1.0436, + "step": 8002 + }, + { + "epoch": 0.962303853784645, + "grad_norm": 3.9745262343500287, + "learning_rate": 1.4843422569380537e-08, + "loss": 1.0551, + "step": 8003 + }, + { + "epoch": 0.9624240966752841, + "grad_norm": 2.0001257618464656, + "learning_rate": 1.4748844489147483e-08, + "loss": 1.0648, + "step": 8004 + }, + { + "epoch": 0.9625443395659231, + "grad_norm": 2.554377079107168, + "learning_rate": 1.4654567569936326e-08, + "loss": 0.9457, + "step": 8005 + }, + { + "epoch": 0.9626645824565623, + "grad_norm": 1.767411939701554, + "learning_rate": 1.456059182604874e-08, + "loss": 1.0675, + "step": 8006 + }, + { + "epoch": 0.9627848253472013, + "grad_norm": 2.2585583151297635, + "learning_rate": 1.4466917271740653e-08, + "loss": 1.004, + "step": 8007 + }, + { + "epoch": 0.9629050682378404, + "grad_norm": 2.13798003132615, + "learning_rate": 1.4373543921222697e-08, + "loss": 0.9094, + "step": 8008 + }, + { + "epoch": 0.9630253111284796, + "grad_norm": 1.93161297692549, + "learning_rate": 1.428047178865932e-08, + "loss": 1.0178, + "step": 8009 + }, + { + "epoch": 0.9631455540191186, + "grad_norm": 1.661055592800267, + "learning_rate": 1.4187700888169451e-08, + "loss": 0.9832, + "step": 8010 + }, + { + "epoch": 0.9632657969097577, + "grad_norm": 0.7583817772186001, + "learning_rate": 1.40952312338265e-08, + "loss": 0.8068, + "step": 8011 + }, + { + "epoch": 0.9633860398003968, + "grad_norm": 1.890693348051118, + "learning_rate": 1.4003062839657909e-08, + "loss": 0.924, + "step": 8012 + }, + { + "epoch": 0.9635062826910359, + "grad_norm": 1.9544796327713965, + "learning_rate": 1.391119571964583e-08, + "loss": 1.0376, + "step": 8013 + }, + { + "epoch": 0.9636265255816749, + "grad_norm": 1.8840245820970716, + "learning_rate": 1.3819629887726225e-08, + "loss": 0.9691, + "step": 8014 + }, + { + "epoch": 0.9637467684723141, + "grad_norm": 2.0802356069346777, + "learning_rate": 1.3728365357789317e-08, + "loss": 0.9996, + "step": 8015 + }, + { + "epoch": 0.9638670113629532, + "grad_norm": 2.666964832167013, + "learning_rate": 1.3637402143680254e-08, + "loss": 1.0011, + "step": 8016 + }, + { + "epoch": 0.9639872542535922, + "grad_norm": 0.7318472273693611, + "learning_rate": 1.3546740259197998e-08, + "loss": 0.7881, + "step": 8017 + }, + { + "epoch": 0.9641074971442314, + "grad_norm": 2.1576044003012838, + "learning_rate": 1.3456379718095989e-08, + "loss": 0.9356, + "step": 8018 + }, + { + "epoch": 0.9642277400348704, + "grad_norm": 0.92261320396483, + "learning_rate": 1.3366320534081487e-08, + "loss": 0.8569, + "step": 8019 + }, + { + "epoch": 0.9643479829255095, + "grad_norm": 2.0604131139937407, + "learning_rate": 1.3276562720816675e-08, + "loss": 0.9945, + "step": 8020 + }, + { + "epoch": 0.9644682258161487, + "grad_norm": 2.8599621112687954, + "learning_rate": 1.3187106291917549e-08, + "loss": 1.063, + "step": 8021 + }, + { + "epoch": 0.9645884687067877, + "grad_norm": 1.9056045815389537, + "learning_rate": 1.309795126095503e-08, + "loss": 0.9393, + "step": 8022 + }, + { + "epoch": 0.9647087115974268, + "grad_norm": 2.5901941030815014, + "learning_rate": 1.3009097641453192e-08, + "loss": 1.0474, + "step": 8023 + }, + { + "epoch": 0.9648289544880659, + "grad_norm": 1.7079059097138656, + "learning_rate": 1.2920545446891474e-08, + "loss": 0.9987, + "step": 8024 + }, + { + "epoch": 0.964949197378705, + "grad_norm": 1.7375969729360274, + "learning_rate": 1.2832294690703127e-08, + "loss": 0.9421, + "step": 8025 + }, + { + "epoch": 0.965069440269344, + "grad_norm": 1.8729785194692732, + "learning_rate": 1.2744345386275668e-08, + "loss": 1.0119, + "step": 8026 + }, + { + "epoch": 0.9651896831599832, + "grad_norm": 1.7526249952812132, + "learning_rate": 1.265669754695109e-08, + "loss": 1.0217, + "step": 8027 + }, + { + "epoch": 0.9653099260506223, + "grad_norm": 2.738627761487715, + "learning_rate": 1.2569351186025201e-08, + "loss": 1.0619, + "step": 8028 + }, + { + "epoch": 0.9654301689412613, + "grad_norm": 1.5732326400722871, + "learning_rate": 1.2482306316748737e-08, + "loss": 0.9922, + "step": 8029 + }, + { + "epoch": 0.9655504118319005, + "grad_norm": 2.153136174985457, + "learning_rate": 1.2395562952326021e-08, + "loss": 1.019, + "step": 8030 + }, + { + "epoch": 0.9656706547225395, + "grad_norm": 2.154746374442103, + "learning_rate": 1.2309121105916309e-08, + "loss": 1.0552, + "step": 8031 + }, + { + "epoch": 0.9657908976131786, + "grad_norm": 2.2496703224252284, + "learning_rate": 1.222298079063222e-08, + "loss": 0.9289, + "step": 8032 + }, + { + "epoch": 0.9659111405038178, + "grad_norm": 2.1520650305866744, + "learning_rate": 1.2137142019541524e-08, + "loss": 0.9692, + "step": 8033 + }, + { + "epoch": 0.9660313833944568, + "grad_norm": 1.9928922715535997, + "learning_rate": 1.2051604805666027e-08, + "loss": 0.975, + "step": 8034 + }, + { + "epoch": 0.9661516262850959, + "grad_norm": 2.3779279830268534, + "learning_rate": 1.196636916198135e-08, + "loss": 1.0196, + "step": 8035 + }, + { + "epoch": 0.9662718691757349, + "grad_norm": 1.9238922573185813, + "learning_rate": 1.1881435101418036e-08, + "loss": 1.0067, + "step": 8036 + }, + { + "epoch": 0.9663921120663741, + "grad_norm": 0.8069543851749693, + "learning_rate": 1.1796802636860003e-08, + "loss": 0.8931, + "step": 8037 + }, + { + "epoch": 0.9665123549570132, + "grad_norm": 2.003280095658767, + "learning_rate": 1.1712471781146316e-08, + "loss": 0.9756, + "step": 8038 + }, + { + "epoch": 0.9666325978476522, + "grad_norm": 1.9945664506937115, + "learning_rate": 1.1628442547069628e-08, + "loss": 0.905, + "step": 8039 + }, + { + "epoch": 0.9667528407382914, + "grad_norm": 2.4203671591431317, + "learning_rate": 1.1544714947377521e-08, + "loss": 1.0092, + "step": 8040 + }, + { + "epoch": 0.9668730836289304, + "grad_norm": 2.0784813138054377, + "learning_rate": 1.1461288994770945e-08, + "loss": 0.9375, + "step": 8041 + }, + { + "epoch": 0.9669933265195695, + "grad_norm": 2.0277513156830635, + "learning_rate": 1.1378164701906002e-08, + "loss": 1.0109, + "step": 8042 + }, + { + "epoch": 0.9671135694102087, + "grad_norm": 1.9173002301613862, + "learning_rate": 1.1295342081392156e-08, + "loss": 0.8994, + "step": 8043 + }, + { + "epoch": 0.9672338123008477, + "grad_norm": 1.7405395925172507, + "learning_rate": 1.1212821145793804e-08, + "loss": 0.9298, + "step": 8044 + }, + { + "epoch": 0.9673540551914868, + "grad_norm": 2.3290030852477956, + "learning_rate": 1.1130601907629156e-08, + "loss": 1.026, + "step": 8045 + }, + { + "epoch": 0.9674742980821259, + "grad_norm": 0.8227542684763669, + "learning_rate": 1.1048684379370899e-08, + "loss": 0.8846, + "step": 8046 + }, + { + "epoch": 0.967594540972765, + "grad_norm": 1.8789510424185907, + "learning_rate": 1.0967068573445759e-08, + "loss": 0.9762, + "step": 8047 + }, + { + "epoch": 0.967714783863404, + "grad_norm": 2.5889384671085964, + "learning_rate": 1.0885754502234945e-08, + "loss": 0.8846, + "step": 8048 + }, + { + "epoch": 0.9678350267540432, + "grad_norm": 4.413940860118764, + "learning_rate": 1.08047421780737e-08, + "loss": 1.0194, + "step": 8049 + }, + { + "epoch": 0.9679552696446823, + "grad_norm": 2.534592670166571, + "learning_rate": 1.0724031613251305e-08, + "loss": 0.9756, + "step": 8050 + }, + { + "epoch": 0.9680755125353213, + "grad_norm": 2.7917472675891184, + "learning_rate": 1.0643622820011744e-08, + "loss": 0.904, + "step": 8051 + }, + { + "epoch": 0.9681957554259605, + "grad_norm": 5.067085920521986, + "learning_rate": 1.0563515810552814e-08, + "loss": 0.916, + "step": 8052 + }, + { + "epoch": 0.9683159983165995, + "grad_norm": 1.5645209403704592, + "learning_rate": 1.0483710597026795e-08, + "loss": 0.9711, + "step": 8053 + }, + { + "epoch": 0.9684362412072386, + "grad_norm": 2.2088552853804004, + "learning_rate": 1.0404207191540227e-08, + "loss": 0.9745, + "step": 8054 + }, + { + "epoch": 0.9685564840978778, + "grad_norm": 2.267429974535621, + "learning_rate": 1.0325005606153236e-08, + "loss": 0.9812, + "step": 8055 + }, + { + "epoch": 0.9686767269885168, + "grad_norm": 2.6438898973898786, + "learning_rate": 1.0246105852881104e-08, + "loss": 1.0285, + "step": 8056 + }, + { + "epoch": 0.9687969698791559, + "grad_norm": 2.8775530550230792, + "learning_rate": 1.0167507943692476e-08, + "loss": 1.0225, + "step": 8057 + }, + { + "epoch": 0.968917212769795, + "grad_norm": 2.1141434636369567, + "learning_rate": 1.008921189051093e-08, + "loss": 0.9495, + "step": 8058 + }, + { + "epoch": 0.9690374556604341, + "grad_norm": 1.970955597174613, + "learning_rate": 1.0011217705213848e-08, + "loss": 1.0083, + "step": 8059 + }, + { + "epoch": 0.9691576985510731, + "grad_norm": 1.8398508772337403, + "learning_rate": 9.933525399632658e-09, + "loss": 0.9804, + "step": 8060 + }, + { + "epoch": 0.9692779414417123, + "grad_norm": 1.6549575369190375, + "learning_rate": 9.856134985553488e-09, + "loss": 0.8899, + "step": 8061 + }, + { + "epoch": 0.9693981843323514, + "grad_norm": 1.5265389978479698, + "learning_rate": 9.77904647471628e-09, + "loss": 0.9713, + "step": 8062 + }, + { + "epoch": 0.9695184272229904, + "grad_norm": 2.1191879900020654, + "learning_rate": 9.702259878815454e-09, + "loss": 0.9753, + "step": 8063 + }, + { + "epoch": 0.9696386701136296, + "grad_norm": 2.7381623115775033, + "learning_rate": 9.625775209499254e-09, + "loss": 0.9792, + "step": 8064 + }, + { + "epoch": 0.9697589130042686, + "grad_norm": 1.989746213573616, + "learning_rate": 9.549592478370172e-09, + "loss": 0.9763, + "step": 8065 + }, + { + "epoch": 0.9698791558949077, + "grad_norm": 1.989091149878206, + "learning_rate": 9.473711696985632e-09, + "loss": 1.0302, + "step": 8066 + }, + { + "epoch": 0.9699993987855468, + "grad_norm": 2.3239593104610217, + "learning_rate": 9.398132876856201e-09, + "loss": 0.9938, + "step": 8067 + }, + { + "epoch": 0.9701196416761859, + "grad_norm": 0.7889285209898497, + "learning_rate": 9.322856029447379e-09, + "loss": 0.8455, + "step": 8068 + }, + { + "epoch": 0.970239884566825, + "grad_norm": 2.0465480164711316, + "learning_rate": 9.247881166178695e-09, + "loss": 1.0408, + "step": 8069 + }, + { + "epoch": 0.970360127457464, + "grad_norm": 2.249498947302713, + "learning_rate": 9.173208298423274e-09, + "loss": 1.0017, + "step": 8070 + }, + { + "epoch": 0.9704803703481032, + "grad_norm": 1.6519445203050536, + "learning_rate": 9.09883743750961e-09, + "loss": 1.0001, + "step": 8071 + }, + { + "epoch": 0.9706006132387422, + "grad_norm": 1.7786397715102673, + "learning_rate": 9.024768594719124e-09, + "loss": 1.0759, + "step": 8072 + }, + { + "epoch": 0.9707208561293813, + "grad_norm": 2.1575234011621984, + "learning_rate": 8.95100178128816e-09, + "loss": 0.9638, + "step": 8073 + }, + { + "epoch": 0.9708410990200205, + "grad_norm": 1.9976688150391892, + "learning_rate": 8.877537008407321e-09, + "loss": 0.9348, + "step": 8074 + }, + { + "epoch": 0.9709613419106595, + "grad_norm": 1.723833231310064, + "learning_rate": 8.804374287221028e-09, + "loss": 0.9232, + "step": 8075 + }, + { + "epoch": 0.9710815848012986, + "grad_norm": 1.6846666922157285, + "learning_rate": 8.731513628827958e-09, + "loss": 1.084, + "step": 8076 + }, + { + "epoch": 0.9712018276919377, + "grad_norm": 2.0521321815318125, + "learning_rate": 8.658955044280825e-09, + "loss": 1.0622, + "step": 8077 + }, + { + "epoch": 0.9713220705825768, + "grad_norm": 1.6671536528879534, + "learning_rate": 8.586698544587268e-09, + "loss": 1.0082, + "step": 8078 + }, + { + "epoch": 0.9714423134732159, + "grad_norm": 2.8456939089697197, + "learning_rate": 8.514744140707853e-09, + "loss": 0.9773, + "step": 8079 + }, + { + "epoch": 0.971562556363855, + "grad_norm": 1.7275053185962115, + "learning_rate": 8.443091843558515e-09, + "loss": 1.0046, + "step": 8080 + }, + { + "epoch": 0.9716827992544941, + "grad_norm": 2.0360257859311512, + "learning_rate": 8.37174166400878e-09, + "loss": 0.8814, + "step": 8081 + }, + { + "epoch": 0.9718030421451331, + "grad_norm": 2.309585590594287, + "learning_rate": 8.300693612881992e-09, + "loss": 1.0906, + "step": 8082 + }, + { + "epoch": 0.9719232850357723, + "grad_norm": 2.004449306023043, + "learning_rate": 8.22994770095664e-09, + "loss": 1.0515, + "step": 8083 + }, + { + "epoch": 0.9720435279264114, + "grad_norm": 2.430912996681807, + "learning_rate": 8.159503938964585e-09, + "loss": 0.9902, + "step": 8084 + }, + { + "epoch": 0.9721637708170504, + "grad_norm": 1.868990996778427, + "learning_rate": 8.089362337592164e-09, + "loss": 0.9397, + "step": 8085 + }, + { + "epoch": 0.9722840137076896, + "grad_norm": 1.644786244102206, + "learning_rate": 8.019522907479536e-09, + "loss": 0.9592, + "step": 8086 + }, + { + "epoch": 0.9724042565983286, + "grad_norm": 2.181559316132543, + "learning_rate": 7.949985659221558e-09, + "loss": 1.0154, + "step": 8087 + }, + { + "epoch": 0.9725244994889677, + "grad_norm": 2.08278804000606, + "learning_rate": 7.880750603366904e-09, + "loss": 1.0262, + "step": 8088 + }, + { + "epoch": 0.9726447423796069, + "grad_norm": 1.921986150241138, + "learning_rate": 7.811817750418282e-09, + "loss": 1.0346, + "step": 8089 + }, + { + "epoch": 0.9727649852702459, + "grad_norm": 1.7418890043037212, + "learning_rate": 7.743187110833105e-09, + "loss": 1.0381, + "step": 8090 + }, + { + "epoch": 0.972885228160885, + "grad_norm": 1.6261264124830517, + "learning_rate": 7.674858695022602e-09, + "loss": 1.0422, + "step": 8091 + }, + { + "epoch": 0.9730054710515241, + "grad_norm": 2.479724490028631, + "learning_rate": 7.606832513351591e-09, + "loss": 0.9942, + "step": 8092 + }, + { + "epoch": 0.9731257139421632, + "grad_norm": 0.8341344479472103, + "learning_rate": 7.539108576140264e-09, + "loss": 0.8768, + "step": 8093 + }, + { + "epoch": 0.9732459568328022, + "grad_norm": 2.450011300041453, + "learning_rate": 7.471686893661732e-09, + "loss": 0.9428, + "step": 8094 + }, + { + "epoch": 0.9733661997234414, + "grad_norm": 1.8106483998891778, + "learning_rate": 7.4045674761442636e-09, + "loss": 0.8804, + "step": 8095 + }, + { + "epoch": 0.9734864426140805, + "grad_norm": 1.7721738640629405, + "learning_rate": 7.337750333769488e-09, + "loss": 0.9805, + "step": 8096 + }, + { + "epoch": 0.9736066855047195, + "grad_norm": 2.217960485837347, + "learning_rate": 7.2712354766737425e-09, + "loss": 0.9641, + "step": 8097 + }, + { + "epoch": 0.9737269283953586, + "grad_norm": 1.7025937292931037, + "learning_rate": 7.2050229149469565e-09, + "loss": 1.0442, + "step": 8098 + }, + { + "epoch": 0.9738471712859977, + "grad_norm": 1.7509195772717856, + "learning_rate": 7.139112658633984e-09, + "loss": 0.872, + "step": 8099 + }, + { + "epoch": 0.9739674141766368, + "grad_norm": 3.8508407814159127, + "learning_rate": 7.073504717733048e-09, + "loss": 0.9415, + "step": 8100 + }, + { + "epoch": 0.9740876570672758, + "grad_norm": 0.742659368590732, + "learning_rate": 7.008199102196855e-09, + "loss": 0.8091, + "step": 8101 + }, + { + "epoch": 0.974207899957915, + "grad_norm": 0.8043931959890694, + "learning_rate": 6.9431958219321464e-09, + "loss": 0.827, + "step": 8102 + }, + { + "epoch": 0.9743281428485541, + "grad_norm": 2.192443928365859, + "learning_rate": 6.878494886800146e-09, + "loss": 1.0152, + "step": 8103 + }, + { + "epoch": 0.9744483857391931, + "grad_norm": 2.2105891623920657, + "learning_rate": 6.814096306615669e-09, + "loss": 0.9983, + "step": 8104 + }, + { + "epoch": 0.9745686286298323, + "grad_norm": 2.3195965108559395, + "learning_rate": 6.750000091148011e-09, + "loss": 0.8872, + "step": 8105 + }, + { + "epoch": 0.9746888715204713, + "grad_norm": 2.2730637075925553, + "learning_rate": 6.686206250120729e-09, + "loss": 0.9598, + "step": 8106 + }, + { + "epoch": 0.9748091144111104, + "grad_norm": 1.9233725690999235, + "learning_rate": 6.622714793210749e-09, + "loss": 0.9817, + "step": 8107 + }, + { + "epoch": 0.9749293573017496, + "grad_norm": 1.6482033947500832, + "learning_rate": 6.559525730050364e-09, + "loss": 1.0215, + "step": 8108 + }, + { + "epoch": 0.9750496001923886, + "grad_norm": 2.1015128090808672, + "learning_rate": 6.496639070224574e-09, + "loss": 0.9978, + "step": 8109 + }, + { + "epoch": 0.9751698430830277, + "grad_norm": 2.1664541091337295, + "learning_rate": 6.4340548232739714e-09, + "loss": 1.0726, + "step": 8110 + }, + { + "epoch": 0.9752900859736668, + "grad_norm": 2.045344355534012, + "learning_rate": 6.371772998692071e-09, + "loss": 1.0283, + "step": 8111 + }, + { + "epoch": 0.9754103288643059, + "grad_norm": 3.2549194370501726, + "learning_rate": 6.309793605927094e-09, + "loss": 0.8871, + "step": 8112 + }, + { + "epoch": 0.975530571754945, + "grad_norm": 2.0369586248751164, + "learning_rate": 6.248116654381297e-09, + "loss": 1.0382, + "step": 8113 + }, + { + "epoch": 0.9756508146455841, + "grad_norm": 1.8139112831720834, + "learning_rate": 6.186742153410751e-09, + "loss": 0.9652, + "step": 8114 + }, + { + "epoch": 0.9757710575362232, + "grad_norm": 2.3291848861536772, + "learning_rate": 6.125670112326453e-09, + "loss": 1.1068, + "step": 8115 + }, + { + "epoch": 0.9758913004268622, + "grad_norm": 1.596189912577689, + "learning_rate": 6.064900540392548e-09, + "loss": 0.937, + "step": 8116 + }, + { + "epoch": 0.9760115433175014, + "grad_norm": 2.297720202165922, + "learning_rate": 6.0044334468278835e-09, + "loss": 1.0229, + "step": 8117 + }, + { + "epoch": 0.9761317862081405, + "grad_norm": 1.7558470185622648, + "learning_rate": 5.944268840805345e-09, + "loss": 0.9507, + "step": 8118 + }, + { + "epoch": 0.9762520290987795, + "grad_norm": 3.3263109304645484, + "learning_rate": 5.88440673145163e-09, + "loss": 0.8767, + "step": 8119 + }, + { + "epoch": 0.9763722719894187, + "grad_norm": 2.2215875023800438, + "learning_rate": 5.824847127848142e-09, + "loss": 1.0626, + "step": 8120 + }, + { + "epoch": 0.9764925148800577, + "grad_norm": 1.8406444839974767, + "learning_rate": 5.765590039029433e-09, + "loss": 1.023, + "step": 8121 + }, + { + "epoch": 0.9766127577706968, + "grad_norm": 8.96180461489573, + "learning_rate": 5.706635473985422e-09, + "loss": 0.9486, + "step": 8122 + }, + { + "epoch": 0.976733000661336, + "grad_norm": 3.803328190988669, + "learning_rate": 5.6479834416591764e-09, + "loss": 1.0894, + "step": 8123 + }, + { + "epoch": 0.976853243551975, + "grad_norm": 1.9679686191584538, + "learning_rate": 5.589633950947803e-09, + "loss": 0.9227, + "step": 8124 + }, + { + "epoch": 0.9769734864426141, + "grad_norm": 2.0879488616615047, + "learning_rate": 5.5315870107035535e-09, + "loss": 0.9321, + "step": 8125 + }, + { + "epoch": 0.9770937293332532, + "grad_norm": 2.093812856157715, + "learning_rate": 5.473842629731607e-09, + "loss": 1.0207, + "step": 8126 + }, + { + "epoch": 0.9772139722238923, + "grad_norm": 1.9617922922171793, + "learning_rate": 5.416400816792066e-09, + "loss": 1.0159, + "step": 8127 + }, + { + "epoch": 0.9773342151145313, + "grad_norm": 2.7592646852174907, + "learning_rate": 5.359261580598407e-09, + "loss": 1.0222, + "step": 8128 + }, + { + "epoch": 0.9774544580051704, + "grad_norm": 2.3969835478115, + "learning_rate": 5.302424929819027e-09, + "loss": 1.0186, + "step": 8129 + }, + { + "epoch": 0.9775747008958096, + "grad_norm": 2.572334676755632, + "learning_rate": 5.24589087307592e-09, + "loss": 0.9648, + "step": 8130 + }, + { + "epoch": 0.9776949437864486, + "grad_norm": 1.527514435969938, + "learning_rate": 5.189659418944891e-09, + "loss": 0.8861, + "step": 8131 + }, + { + "epoch": 0.9778151866770877, + "grad_norm": 2.0824575363537554, + "learning_rate": 5.133730575956674e-09, + "loss": 1.0204, + "step": 8132 + }, + { + "epoch": 0.9779354295677268, + "grad_norm": 2.268603452042606, + "learning_rate": 5.0781043525953696e-09, + "loss": 0.9592, + "step": 8133 + }, + { + "epoch": 0.9780556724583659, + "grad_norm": 1.8831770927752112, + "learning_rate": 5.0227807572995605e-09, + "loss": 0.9777, + "step": 8134 + }, + { + "epoch": 0.9781759153490049, + "grad_norm": 1.9879473272787047, + "learning_rate": 4.967759798461646e-09, + "loss": 0.9127, + "step": 8135 + }, + { + "epoch": 0.9782961582396441, + "grad_norm": 1.981886136430617, + "learning_rate": 4.913041484428282e-09, + "loss": 0.9859, + "step": 8136 + }, + { + "epoch": 0.9784164011302832, + "grad_norm": 3.119589585184672, + "learning_rate": 4.858625823500384e-09, + "loss": 0.9814, + "step": 8137 + }, + { + "epoch": 0.9785366440209222, + "grad_norm": 2.4354389965207353, + "learning_rate": 4.80451282393246e-09, + "loss": 0.9697, + "step": 8138 + }, + { + "epoch": 0.9786568869115614, + "grad_norm": 3.2112959687182454, + "learning_rate": 4.750702493933722e-09, + "loss": 0.9111, + "step": 8139 + }, + { + "epoch": 0.9787771298022004, + "grad_norm": 2.4762699183010057, + "learning_rate": 4.697194841666974e-09, + "loss": 1.0871, + "step": 8140 + }, + { + "epoch": 0.9788973726928395, + "grad_norm": 1.7921257613705215, + "learning_rate": 4.6439898752492764e-09, + "loss": 1.0566, + "step": 8141 + }, + { + "epoch": 0.9790176155834787, + "grad_norm": 0.7645730167770447, + "learning_rate": 4.591087602751731e-09, + "loss": 0.8743, + "step": 8142 + }, + { + "epoch": 0.9791378584741177, + "grad_norm": 1.8881529733302334, + "learning_rate": 4.538488032199916e-09, + "loss": 0.9577, + "step": 8143 + }, + { + "epoch": 0.9792581013647568, + "grad_norm": 2.0251624807469875, + "learning_rate": 4.486191171572784e-09, + "loss": 0.9282, + "step": 8144 + }, + { + "epoch": 0.9793783442553959, + "grad_norm": 1.7259212355958522, + "learning_rate": 4.434197028803766e-09, + "loss": 1.0135, + "step": 8145 + }, + { + "epoch": 0.979498587146035, + "grad_norm": 2.17187368269407, + "learning_rate": 4.3825056117805514e-09, + "loss": 1.0551, + "step": 8146 + }, + { + "epoch": 0.979618830036674, + "grad_norm": 3.0610793574956108, + "learning_rate": 4.331116928344425e-09, + "loss": 1.0286, + "step": 8147 + }, + { + "epoch": 0.9797390729273132, + "grad_norm": 1.9838474804027875, + "learning_rate": 4.28003098629115e-09, + "loss": 0.8624, + "step": 8148 + }, + { + "epoch": 0.9798593158179523, + "grad_norm": 2.226946024740369, + "learning_rate": 4.229247793370305e-09, + "loss": 1.0239, + "step": 8149 + }, + { + "epoch": 0.9799795587085913, + "grad_norm": 1.5957655036528129, + "learning_rate": 4.178767357285951e-09, + "loss": 0.9388, + "step": 8150 + }, + { + "epoch": 0.9800998015992305, + "grad_norm": 1.908383910440372, + "learning_rate": 4.128589685695516e-09, + "loss": 0.951, + "step": 8151 + }, + { + "epoch": 0.9802200444898695, + "grad_norm": 2.190142751770378, + "learning_rate": 4.078714786211135e-09, + "loss": 1.0817, + "step": 8152 + }, + { + "epoch": 0.9803402873805086, + "grad_norm": 1.8119521793008893, + "learning_rate": 4.029142666398977e-09, + "loss": 1.0041, + "step": 8153 + }, + { + "epoch": 0.9804605302711478, + "grad_norm": 1.8922418033736312, + "learning_rate": 3.979873333778805e-09, + "loss": 1.0348, + "step": 8154 + }, + { + "epoch": 0.9805807731617868, + "grad_norm": 3.387683489007185, + "learning_rate": 3.930906795824862e-09, + "loss": 0.9773, + "step": 8155 + }, + { + "epoch": 0.9807010160524259, + "grad_norm": 2.086824721341622, + "learning_rate": 3.882243059965207e-09, + "loss": 1.0048, + "step": 8156 + }, + { + "epoch": 0.980821258943065, + "grad_norm": 2.922960701844342, + "learning_rate": 3.833882133582156e-09, + "loss": 0.8958, + "step": 8157 + }, + { + "epoch": 0.9809415018337041, + "grad_norm": 1.628257420988434, + "learning_rate": 3.785824024012285e-09, + "loss": 1.0207, + "step": 8158 + }, + { + "epoch": 0.9810617447243432, + "grad_norm": 1.5903475476340463, + "learning_rate": 3.738068738545541e-09, + "loss": 1.0202, + "step": 8159 + }, + { + "epoch": 0.9811819876149822, + "grad_norm": 2.2019867899028127, + "learning_rate": 3.6906162844265733e-09, + "loss": 1.0247, + "step": 8160 + }, + { + "epoch": 0.9813022305056214, + "grad_norm": 2.4670479617240475, + "learning_rate": 3.643466668853845e-09, + "loss": 0.9465, + "step": 8161 + }, + { + "epoch": 0.9814224733962604, + "grad_norm": 2.0935213334203997, + "learning_rate": 3.59661989898008e-09, + "loss": 0.988, + "step": 8162 + }, + { + "epoch": 0.9815427162868995, + "grad_norm": 2.2673832877844817, + "learning_rate": 3.5500759819115934e-09, + "loss": 1.0066, + "step": 8163 + }, + { + "epoch": 0.9816629591775387, + "grad_norm": 2.2511606401821425, + "learning_rate": 3.5038349247094034e-09, + "loss": 1.0475, + "step": 8164 + }, + { + "epoch": 0.9817832020681777, + "grad_norm": 2.1736900374843784, + "learning_rate": 3.4578967343878994e-09, + "loss": 1.0114, + "step": 8165 + }, + { + "epoch": 0.9819034449588168, + "grad_norm": 2.1811206986979923, + "learning_rate": 3.4122614179161733e-09, + "loss": 1.049, + "step": 8166 + }, + { + "epoch": 0.9820236878494559, + "grad_norm": 1.7938957394399244, + "learning_rate": 3.36692898221691e-09, + "loss": 1.0184, + "step": 8167 + }, + { + "epoch": 0.982143930740095, + "grad_norm": 1.790989400324294, + "learning_rate": 3.3218994341668305e-09, + "loss": 0.9703, + "step": 8168 + }, + { + "epoch": 0.982264173630734, + "grad_norm": 1.826349880090402, + "learning_rate": 3.2771727805971373e-09, + "loss": 0.9918, + "step": 8169 + }, + { + "epoch": 0.9823844165213732, + "grad_norm": 1.7444096237130584, + "learning_rate": 3.232749028292847e-09, + "loss": 1.0101, + "step": 8170 + }, + { + "epoch": 0.9825046594120123, + "grad_norm": 2.008803542348761, + "learning_rate": 3.188628183992792e-09, + "loss": 1.1192, + "step": 8171 + }, + { + "epoch": 0.9826249023026513, + "grad_norm": 0.7458590792540183, + "learning_rate": 3.1448102543902844e-09, + "loss": 0.8632, + "step": 8172 + }, + { + "epoch": 0.9827451451932905, + "grad_norm": 1.974901066487157, + "learning_rate": 3.1012952461324515e-09, + "loss": 0.9106, + "step": 8173 + }, + { + "epoch": 0.9828653880839295, + "grad_norm": 2.9485488231770374, + "learning_rate": 3.0580831658204575e-09, + "loss": 0.9771, + "step": 8174 + }, + { + "epoch": 0.9829856309745686, + "grad_norm": 1.7154683081496636, + "learning_rate": 3.015174020009281e-09, + "loss": 1.0172, + "step": 8175 + }, + { + "epoch": 0.9831058738652078, + "grad_norm": 1.853498774332321, + "learning_rate": 2.9725678152086043e-09, + "loss": 0.9872, + "step": 8176 + }, + { + "epoch": 0.9832261167558468, + "grad_norm": 2.8138150001993165, + "learning_rate": 2.930264557881257e-09, + "loss": 1.0592, + "step": 8177 + }, + { + "epoch": 0.9833463596464859, + "grad_norm": 0.8706547952607443, + "learning_rate": 2.8882642544452163e-09, + "loss": 0.8193, + "step": 8178 + }, + { + "epoch": 0.983466602537125, + "grad_norm": 2.208035670918351, + "learning_rate": 2.8465669112716083e-09, + "loss": 0.9778, + "step": 8179 + }, + { + "epoch": 0.9835868454277641, + "grad_norm": 1.8674323268614326, + "learning_rate": 2.8051725346858177e-09, + "loss": 1.0004, + "step": 8180 + }, + { + "epoch": 0.9837070883184031, + "grad_norm": 2.2331641057175324, + "learning_rate": 2.7640811309674883e-09, + "loss": 0.9483, + "step": 8181 + }, + { + "epoch": 0.9838273312090423, + "grad_norm": 1.6407458306276588, + "learning_rate": 2.7232927063498557e-09, + "loss": 1.0471, + "step": 8182 + }, + { + "epoch": 0.9839475740996814, + "grad_norm": 1.717575573653616, + "learning_rate": 2.682807267020859e-09, + "loss": 0.9261, + "step": 8183 + }, + { + "epoch": 0.9840678169903204, + "grad_norm": 1.8391720462865009, + "learning_rate": 2.642624819121808e-09, + "loss": 0.8602, + "step": 8184 + }, + { + "epoch": 0.9841880598809596, + "grad_norm": 1.898538092314708, + "learning_rate": 2.6027453687487154e-09, + "loss": 0.8535, + "step": 8185 + }, + { + "epoch": 0.9843083027715986, + "grad_norm": 2.6104308407807917, + "learning_rate": 2.5631689219509643e-09, + "loss": 0.7773, + "step": 8186 + }, + { + "epoch": 0.9844285456622377, + "grad_norm": 2.4240401895294768, + "learning_rate": 2.523895484732197e-09, + "loss": 1.0715, + "step": 8187 + }, + { + "epoch": 0.9845487885528769, + "grad_norm": 1.9964723470613277, + "learning_rate": 2.4849250630505357e-09, + "loss": 0.9836, + "step": 8188 + }, + { + "epoch": 0.9846690314435159, + "grad_norm": 2.4172343431408185, + "learning_rate": 2.4462576628172528e-09, + "loss": 0.9685, + "step": 8189 + }, + { + "epoch": 0.984789274334155, + "grad_norm": 1.9438192618930634, + "learning_rate": 2.407893289898766e-09, + "loss": 0.9718, + "step": 8190 + }, + { + "epoch": 0.984909517224794, + "grad_norm": 1.8467441105813878, + "learning_rate": 2.3698319501144202e-09, + "loss": 1.0746, + "step": 8191 + }, + { + "epoch": 0.9850297601154332, + "grad_norm": 1.911969032631369, + "learning_rate": 2.3320736492382644e-09, + "loss": 0.9718, + "step": 8192 + }, + { + "epoch": 0.9851500030060723, + "grad_norm": 3.4320524573240747, + "learning_rate": 2.29461839299816e-09, + "loss": 0.9164, + "step": 8193 + }, + { + "epoch": 0.9852702458967113, + "grad_norm": 1.6081390820274735, + "learning_rate": 2.257466187076229e-09, + "loss": 1.0368, + "step": 8194 + }, + { + "epoch": 0.9853904887873505, + "grad_norm": 1.7701854044045209, + "learning_rate": 2.2206170371081854e-09, + "loss": 0.9479, + "step": 8195 + }, + { + "epoch": 0.9855107316779895, + "grad_norm": 1.6566679804980302, + "learning_rate": 2.1840709486842247e-09, + "loss": 1.0848, + "step": 8196 + }, + { + "epoch": 0.9856309745686286, + "grad_norm": 2.216574448505306, + "learning_rate": 2.1478279273481335e-09, + "loss": 1.0254, + "step": 8197 + }, + { + "epoch": 0.9857512174592677, + "grad_norm": 3.607370852288739, + "learning_rate": 2.1118879785981815e-09, + "loss": 1.0337, + "step": 8198 + }, + { + "epoch": 0.9858714603499068, + "grad_norm": 2.2039245885973777, + "learning_rate": 2.0762511078862288e-09, + "loss": 1.0302, + "step": 8199 + }, + { + "epoch": 0.9859917032405459, + "grad_norm": 1.9242406412506399, + "learning_rate": 2.0409173206186183e-09, + "loss": 0.8856, + "step": 8200 + }, + { + "epoch": 0.986111946131185, + "grad_norm": 2.339464261927699, + "learning_rate": 2.0058866221550617e-09, + "loss": 1.1101, + "step": 8201 + }, + { + "epoch": 0.9862321890218241, + "grad_norm": 2.1177468774993398, + "learning_rate": 1.971159017809976e-09, + "loss": 0.9905, + "step": 8202 + }, + { + "epoch": 0.9863524319124631, + "grad_norm": 2.0992473343385574, + "learning_rate": 1.93673451285159e-09, + "loss": 1.0178, + "step": 8203 + }, + { + "epoch": 0.9864726748031023, + "grad_norm": 0.7497194480687629, + "learning_rate": 1.9026131125019495e-09, + "loss": 0.8033, + "step": 8204 + }, + { + "epoch": 0.9865929176937414, + "grad_norm": 1.7370816667121303, + "learning_rate": 1.8687948219371363e-09, + "loss": 1.1086, + "step": 8205 + }, + { + "epoch": 0.9867131605843804, + "grad_norm": 2.150986148096207, + "learning_rate": 1.835279646287491e-09, + "loss": 1.1253, + "step": 8206 + }, + { + "epoch": 0.9868334034750196, + "grad_norm": 1.824498810600213, + "learning_rate": 1.8020675906371685e-09, + "loss": 1.0004, + "step": 8207 + }, + { + "epoch": 0.9869536463656586, + "grad_norm": 1.9579623324089885, + "learning_rate": 1.7691586600243612e-09, + "loss": 0.9869, + "step": 8208 + }, + { + "epoch": 0.9870738892562977, + "grad_norm": 2.3994476477477535, + "learning_rate": 1.7365528594415202e-09, + "loss": 1.1046, + "step": 8209 + }, + { + "epoch": 0.9871941321469369, + "grad_norm": 1.9115193190101043, + "learning_rate": 1.7042501938346888e-09, + "loss": 0.9136, + "step": 8210 + }, + { + "epoch": 0.9873143750375759, + "grad_norm": 2.368377487161178, + "learning_rate": 1.6722506681043913e-09, + "loss": 1.0009, + "step": 8211 + }, + { + "epoch": 0.987434617928215, + "grad_norm": 2.3575001524260744, + "learning_rate": 1.640554287104745e-09, + "loss": 0.9296, + "step": 8212 + }, + { + "epoch": 0.9875548608188541, + "grad_norm": 2.0613988769140317, + "learning_rate": 1.609161055644348e-09, + "loss": 1.0364, + "step": 8213 + }, + { + "epoch": 0.9876751037094932, + "grad_norm": 2.168673563992664, + "learning_rate": 1.5780709784849467e-09, + "loss": 0.9262, + "step": 8214 + }, + { + "epoch": 0.9877953466001322, + "grad_norm": 2.0419440827069324, + "learning_rate": 1.5472840603436565e-09, + "loss": 1.0587, + "step": 8215 + }, + { + "epoch": 0.9879155894907714, + "grad_norm": 1.992112700610177, + "learning_rate": 1.5168003058900757e-09, + "loss": 1.0176, + "step": 8216 + }, + { + "epoch": 0.9880358323814105, + "grad_norm": 1.8781653706151789, + "learning_rate": 1.4866197197491715e-09, + "loss": 1.1567, + "step": 8217 + }, + { + "epoch": 0.9881560752720495, + "grad_norm": 3.4940532846127135, + "learning_rate": 1.4567423064988371e-09, + "loss": 1.0183, + "step": 8218 + }, + { + "epoch": 0.9882763181626887, + "grad_norm": 2.0725043444248583, + "learning_rate": 1.4271680706718913e-09, + "loss": 1.0158, + "step": 8219 + }, + { + "epoch": 0.9883965610533277, + "grad_norm": 2.234953585319824, + "learning_rate": 1.3978970167543013e-09, + "loss": 1.0661, + "step": 8220 + }, + { + "epoch": 0.9885168039439668, + "grad_norm": 2.3540037543267687, + "learning_rate": 1.3689291491867372e-09, + "loss": 1.0105, + "step": 8221 + }, + { + "epoch": 0.988637046834606, + "grad_norm": 2.050365006120909, + "learning_rate": 1.3402644723636836e-09, + "loss": 0.9746, + "step": 8222 + }, + { + "epoch": 0.988757289725245, + "grad_norm": 2.2870738273483746, + "learning_rate": 1.311902990633218e-09, + "loss": 1.0721, + "step": 8223 + }, + { + "epoch": 0.9888775326158841, + "grad_norm": 1.8853945300522132, + "learning_rate": 1.2838447082978987e-09, + "loss": 0.9515, + "step": 8224 + }, + { + "epoch": 0.9889977755065231, + "grad_norm": 3.1755034640771864, + "learning_rate": 1.2560896296143208e-09, + "loss": 1.0687, + "step": 8225 + }, + { + "epoch": 0.9891180183971623, + "grad_norm": 2.9544887724020885, + "learning_rate": 1.2286377587926722e-09, + "loss": 1.0631, + "step": 8226 + }, + { + "epoch": 0.9892382612878013, + "grad_norm": 2.0802192174492724, + "learning_rate": 1.2014890999973992e-09, + "loss": 0.9838, + "step": 8227 + }, + { + "epoch": 0.9893585041784404, + "grad_norm": 1.6563306824807578, + "learning_rate": 1.1746436573472073e-09, + "loss": 1.0186, + "step": 8228 + }, + { + "epoch": 0.9894787470690796, + "grad_norm": 2.0762026507302735, + "learning_rate": 1.1481014349141726e-09, + "loss": 0.9284, + "step": 8229 + }, + { + "epoch": 0.9895989899597186, + "grad_norm": 3.73244180479373, + "learning_rate": 1.121862436724852e-09, + "loss": 1.0849, + "step": 8230 + }, + { + "epoch": 0.9897192328503577, + "grad_norm": 1.82537099646887, + "learning_rate": 1.0959266667598388e-09, + "loss": 0.9386, + "step": 8231 + }, + { + "epoch": 0.9898394757409968, + "grad_norm": 3.9932232910898486, + "learning_rate": 1.0702941289533196e-09, + "loss": 0.9844, + "step": 8232 + }, + { + "epoch": 0.9899597186316359, + "grad_norm": 2.0490885406324137, + "learning_rate": 1.0449648271939615e-09, + "loss": 1.1208, + "step": 8233 + }, + { + "epoch": 0.990079961522275, + "grad_norm": 1.559791506259496, + "learning_rate": 1.0199387653240243e-09, + "loss": 0.962, + "step": 8234 + }, + { + "epoch": 0.9902002044129141, + "grad_norm": 1.8312234297247165, + "learning_rate": 9.952159471400267e-10, + "loss": 0.9472, + "step": 8235 + }, + { + "epoch": 0.9903204473035532, + "grad_norm": 1.9016312473129053, + "learning_rate": 9.707963763923022e-10, + "loss": 1.0787, + "step": 8236 + }, + { + "epoch": 0.9904406901941922, + "grad_norm": 1.8090206820527184, + "learning_rate": 9.466800567854427e-10, + "loss": 1.0262, + "step": 8237 + }, + { + "epoch": 0.9905609330848314, + "grad_norm": 1.8909987403385318, + "learning_rate": 9.228669919778553e-10, + "loss": 0.9189, + "step": 8238 + }, + { + "epoch": 0.9906811759754705, + "grad_norm": 2.247635870911752, + "learning_rate": 8.993571855817617e-10, + "loss": 1.0284, + "step": 8239 + }, + { + "epoch": 0.9908014188661095, + "grad_norm": 2.032208260669631, + "learning_rate": 8.761506411638642e-10, + "loss": 0.9846, + "step": 8240 + }, + { + "epoch": 0.9909216617567487, + "grad_norm": 1.9320501157685606, + "learning_rate": 8.53247362244236e-10, + "loss": 0.974, + "step": 8241 + }, + { + "epoch": 0.9910419046473877, + "grad_norm": 2.314923434216912, + "learning_rate": 8.306473522976532e-10, + "loss": 0.917, + "step": 8242 + }, + { + "epoch": 0.9911621475380268, + "grad_norm": 1.7834190413417323, + "learning_rate": 8.083506147522623e-10, + "loss": 0.9503, + "step": 8243 + }, + { + "epoch": 0.991282390428666, + "grad_norm": 2.507746900609607, + "learning_rate": 7.863571529906909e-10, + "loss": 1.0914, + "step": 8244 + }, + { + "epoch": 0.991402633319305, + "grad_norm": 0.7368065370135093, + "learning_rate": 7.646669703489372e-10, + "loss": 0.8341, + "step": 8245 + }, + { + "epoch": 0.9915228762099441, + "grad_norm": 1.7747580699186392, + "learning_rate": 7.432800701177023e-10, + "loss": 0.81, + "step": 8246 + }, + { + "epoch": 0.9916431191005832, + "grad_norm": 0.8073263140390851, + "learning_rate": 7.221964555415017e-10, + "loss": 0.8149, + "step": 8247 + }, + { + "epoch": 0.9917633619912223, + "grad_norm": 1.8613059011535604, + "learning_rate": 7.01416129818222e-10, + "loss": 0.9866, + "step": 8248 + }, + { + "epoch": 0.9918836048818613, + "grad_norm": 2.0609160612320174, + "learning_rate": 6.809390961006745e-10, + "loss": 0.8232, + "step": 8249 + }, + { + "epoch": 0.9920038477725005, + "grad_norm": 2.008739094351485, + "learning_rate": 6.607653574948191e-10, + "loss": 0.9245, + "step": 8250 + }, + { + "epoch": 0.9921240906631396, + "grad_norm": 3.7761779059025233, + "learning_rate": 6.408949170613187e-10, + "loss": 1.0536, + "step": 8251 + }, + { + "epoch": 0.9922443335537786, + "grad_norm": 1.6501336072522876, + "learning_rate": 6.213277778144288e-10, + "loss": 1.0539, + "step": 8252 + }, + { + "epoch": 0.9923645764444178, + "grad_norm": 2.3011416297107385, + "learning_rate": 6.020639427224416e-10, + "loss": 0.9031, + "step": 8253 + }, + { + "epoch": 0.9924848193350568, + "grad_norm": 2.1783066617718743, + "learning_rate": 5.831034147076864e-10, + "loss": 0.9653, + "step": 8254 + }, + { + "epoch": 0.9926050622256959, + "grad_norm": 0.7071714314966955, + "learning_rate": 5.644461966463065e-10, + "loss": 0.7953, + "step": 8255 + }, + { + "epoch": 0.9927253051163349, + "grad_norm": 3.2141824554327223, + "learning_rate": 5.460922913687049e-10, + "loss": 0.996, + "step": 8256 + }, + { + "epoch": 0.9928455480069741, + "grad_norm": 2.050010318264929, + "learning_rate": 5.280417016593208e-10, + "loss": 0.9891, + "step": 8257 + }, + { + "epoch": 0.9929657908976132, + "grad_norm": 1.7487237908974036, + "learning_rate": 5.102944302559642e-10, + "loss": 0.9859, + "step": 8258 + }, + { + "epoch": 0.9930860337882522, + "grad_norm": 1.8882216633938567, + "learning_rate": 4.9285047985137e-10, + "loss": 1.0351, + "step": 8259 + }, + { + "epoch": 0.9932062766788914, + "grad_norm": 1.673663599748555, + "learning_rate": 4.757098530916436e-10, + "loss": 0.9825, + "step": 8260 + }, + { + "epoch": 0.9933265195695304, + "grad_norm": 5.176956676671044, + "learning_rate": 4.5887255257670563e-10, + "loss": 1.0099, + "step": 8261 + }, + { + "epoch": 0.9934467624601695, + "grad_norm": 2.138426828498531, + "learning_rate": 4.4233858086117906e-10, + "loss": 1.006, + "step": 8262 + }, + { + "epoch": 0.9935670053508087, + "grad_norm": 2.386235214675171, + "learning_rate": 4.261079404528356e-10, + "loss": 0.9144, + "step": 8263 + }, + { + "epoch": 0.9936872482414477, + "grad_norm": 2.2193587636608867, + "learning_rate": 4.1018063381437205e-10, + "loss": 0.9276, + "step": 8264 + }, + { + "epoch": 0.9938074911320868, + "grad_norm": 0.8706303890852901, + "learning_rate": 3.9455666336141167e-10, + "loss": 0.8484, + "step": 8265 + }, + { + "epoch": 0.9939277340227259, + "grad_norm": 3.0990775120148375, + "learning_rate": 3.7923603146450267e-10, + "loss": 1.0636, + "step": 8266 + }, + { + "epoch": 0.994047976913365, + "grad_norm": 2.0403257235242984, + "learning_rate": 3.642187404473418e-10, + "loss": 1.0457, + "step": 8267 + }, + { + "epoch": 0.994168219804004, + "grad_norm": 2.1154991258958864, + "learning_rate": 3.495047925885508e-10, + "loss": 1.0959, + "step": 8268 + }, + { + "epoch": 0.9942884626946432, + "grad_norm": 2.650036624439436, + "learning_rate": 3.350941901199e-10, + "loss": 1.0634, + "step": 8269 + }, + { + "epoch": 0.9944087055852823, + "grad_norm": 3.108700933188073, + "learning_rate": 3.2098693522764066e-10, + "loss": 1.07, + "step": 8270 + }, + { + "epoch": 0.9945289484759213, + "grad_norm": 1.876883577929955, + "learning_rate": 3.071830300516165e-10, + "loss": 1.0457, + "step": 8271 + }, + { + "epoch": 0.9946491913665605, + "grad_norm": 2.1647110941402206, + "learning_rate": 2.9368247668615234e-10, + "loss": 0.9442, + "step": 8272 + }, + { + "epoch": 0.9947694342571995, + "grad_norm": 2.3336690138999505, + "learning_rate": 2.804852771789434e-10, + "loss": 0.8499, + "step": 8273 + }, + { + "epoch": 0.9948896771478386, + "grad_norm": 1.7764240274673047, + "learning_rate": 2.675914335321661e-10, + "loss": 0.7926, + "step": 8274 + }, + { + "epoch": 0.9950099200384778, + "grad_norm": 2.680664507200555, + "learning_rate": 2.550009477018111e-10, + "loss": 1.031, + "step": 8275 + }, + { + "epoch": 0.9951301629291168, + "grad_norm": 2.067430210065731, + "learning_rate": 2.4271382159790634e-10, + "loss": 0.8691, + "step": 8276 + }, + { + "epoch": 0.9952504058197559, + "grad_norm": 1.8401780523314113, + "learning_rate": 2.3073005708429406e-10, + "loss": 1.0962, + "step": 8277 + }, + { + "epoch": 0.995370648710395, + "grad_norm": 1.7608823292034665, + "learning_rate": 2.190496559788535e-10, + "loss": 0.9595, + "step": 8278 + }, + { + "epoch": 0.9954908916010341, + "grad_norm": 2.623954837358081, + "learning_rate": 2.0767262005372265e-10, + "loss": 1.0034, + "step": 8279 + }, + { + "epoch": 0.9956111344916732, + "grad_norm": 1.9629932585497007, + "learning_rate": 1.965989510346322e-10, + "loss": 0.9902, + "step": 8280 + }, + { + "epoch": 0.9957313773823123, + "grad_norm": 2.0353284532322102, + "learning_rate": 1.8582865060134955e-10, + "loss": 0.948, + "step": 8281 + }, + { + "epoch": 0.9958516202729514, + "grad_norm": 0.7837660899480414, + "learning_rate": 1.7536172038790098e-10, + "loss": 0.7947, + "step": 8282 + }, + { + "epoch": 0.9959718631635904, + "grad_norm": 2.375797740072458, + "learning_rate": 1.651981619819054e-10, + "loss": 0.9317, + "step": 8283 + }, + { + "epoch": 0.9960921060542296, + "grad_norm": 3.1726241706855176, + "learning_rate": 1.5533797692546257e-10, + "loss": 0.9407, + "step": 8284 + }, + { + "epoch": 0.9962123489448687, + "grad_norm": 2.1352479325286606, + "learning_rate": 1.4578116671404296e-10, + "loss": 1.0764, + "step": 8285 + }, + { + "epoch": 0.9963325918355077, + "grad_norm": 1.9543594731567124, + "learning_rate": 1.3652773279759777e-10, + "loss": 0.9472, + "step": 8286 + }, + { + "epoch": 0.9964528347261468, + "grad_norm": 1.6144586658544224, + "learning_rate": 1.2757767657989305e-10, + "loss": 0.8624, + "step": 8287 + }, + { + "epoch": 0.9965730776167859, + "grad_norm": 2.6690584026877358, + "learning_rate": 1.1893099941850948e-10, + "loss": 1.1047, + "step": 8288 + }, + { + "epoch": 0.996693320507425, + "grad_norm": 2.180102783894821, + "learning_rate": 1.105877026252866e-10, + "loss": 1.0117, + "step": 8289 + }, + { + "epoch": 0.996813563398064, + "grad_norm": 2.6469191978754822, + "learning_rate": 1.0254778746565663e-10, + "loss": 0.9579, + "step": 8290 + }, + { + "epoch": 0.9969338062887032, + "grad_norm": 2.070524384794186, + "learning_rate": 9.481125515953259e-11, + "loss": 0.9683, + "step": 8291 + }, + { + "epoch": 0.9970540491793423, + "grad_norm": 2.0099914839075463, + "learning_rate": 8.737810688064228e-11, + "loss": 1.0328, + "step": 8292 + }, + { + "epoch": 0.9971742920699813, + "grad_norm": 1.9989979140865708, + "learning_rate": 8.024834375608414e-11, + "loss": 1.0289, + "step": 8293 + }, + { + "epoch": 0.9972945349606205, + "grad_norm": 0.8829975103103428, + "learning_rate": 7.342196686788149e-11, + "loss": 0.8659, + "step": 8294 + }, + { + "epoch": 0.9974147778512595, + "grad_norm": 2.1050721539886785, + "learning_rate": 6.689897725142834e-11, + "loss": 0.9215, + "step": 8295 + }, + { + "epoch": 0.9975350207418986, + "grad_norm": 2.7646895728626846, + "learning_rate": 6.067937589615545e-11, + "loss": 1.1209, + "step": 8296 + }, + { + "epoch": 0.9976552636325378, + "grad_norm": 0.7773851163199454, + "learning_rate": 5.476316374575241e-11, + "loss": 0.8142, + "step": 8297 + }, + { + "epoch": 0.9977755065231768, + "grad_norm": 2.0568092907378595, + "learning_rate": 4.9150341697723476e-11, + "loss": 0.9594, + "step": 8298 + }, + { + "epoch": 0.9978957494138159, + "grad_norm": 1.716539105364087, + "learning_rate": 4.384091060338768e-11, + "loss": 0.904, + "step": 8299 + }, + { + "epoch": 0.998015992304455, + "grad_norm": 2.203861302915084, + "learning_rate": 3.883487126810081e-11, + "loss": 0.9761, + "step": 8300 + }, + { + "epoch": 0.9981362351950941, + "grad_norm": 1.7477126158900063, + "learning_rate": 3.41322244516995e-11, + "loss": 1.0323, + "step": 8301 + }, + { + "epoch": 0.9982564780857331, + "grad_norm": 1.817661368593131, + "learning_rate": 2.9732970866946925e-11, + "loss": 0.8675, + "step": 8302 + }, + { + "epoch": 0.9983767209763723, + "grad_norm": 2.7977043073159673, + "learning_rate": 2.563711118175327e-11, + "loss": 1.0126, + "step": 8303 + }, + { + "epoch": 0.9984969638670114, + "grad_norm": 1.8125881305183469, + "learning_rate": 2.184464601717728e-11, + "loss": 1.0745, + "step": 8304 + }, + { + "epoch": 0.9986172067576504, + "grad_norm": 4.0306852669631095, + "learning_rate": 1.8355575948758585e-11, + "loss": 1.0114, + "step": 8305 + }, + { + "epoch": 0.9987374496482896, + "grad_norm": 2.1309954270269484, + "learning_rate": 1.5169901505407424e-11, + "loss": 0.9701, + "step": 8306 + }, + { + "epoch": 0.9988576925389286, + "grad_norm": 1.6962230012863206, + "learning_rate": 1.228762317073695e-11, + "loss": 0.9757, + "step": 8307 + }, + { + "epoch": 0.9989779354295677, + "grad_norm": 2.0857431842720486, + "learning_rate": 9.70874138195299e-12, + "loss": 1.0282, + "step": 8308 + }, + { + "epoch": 0.9990981783202069, + "grad_norm": 1.540809298434974, + "learning_rate": 7.433256530076093e-12, + "loss": 0.9832, + "step": 8309 + }, + { + "epoch": 0.9992184212108459, + "grad_norm": 2.0454641823589768, + "learning_rate": 5.46116896038562e-12, + "loss": 0.993, + "step": 8310 + }, + { + "epoch": 0.999338664101485, + "grad_norm": 2.180149043176605, + "learning_rate": 3.792478972197699e-12, + "loss": 0.8575, + "step": 8311 + }, + { + "epoch": 0.9994589069921241, + "grad_norm": 2.1640241598871293, + "learning_rate": 2.4271868181990895e-12, + "loss": 0.9343, + "step": 8312 + }, + { + "epoch": 0.9995791498827632, + "grad_norm": 2.073242417495866, + "learning_rate": 1.3652927060014973e-12, + "loss": 1.0479, + "step": 8313 + }, + { + "epoch": 0.9996993927734023, + "grad_norm": 2.110144580750893, + "learning_rate": 6.067967965872612e-13, + "loss": 0.8777, + "step": 8314 + }, + { + "epoch": 0.9998196356640414, + "grad_norm": 1.6961684637073615, + "learning_rate": 1.5169920497548615e-13, + "loss": 1.0071, + "step": 8315 + }, + { + "epoch": 0.9999398785546805, + "grad_norm": 1.4517307771146142, + "learning_rate": 0.0, + "loss": 0.788, + "step": 8316 + }, + { + "epoch": 0.9999398785546805, + "step": 8316, + "total_flos": 6.768746676963246e+17, + "train_loss": 1.0048440673019865, + "train_runtime": 114481.4061, + "train_samples_per_second": 2.906, + "train_steps_per_second": 0.073 + } + ], + "logging_steps": 1.0, + "max_steps": 8316, + "num_input_tokens_seen": 0, + "num_train_epochs": 1, + "save_steps": 100, + "stateful_callbacks": { + "TrainerControl": { + "args": { + "should_epoch_stop": false, + "should_evaluate": false, + "should_log": false, + "should_save": true, + "should_training_stop": true + }, + "attributes": {} + } + }, + "total_flos": 6.768746676963246e+17, + "train_batch_size": 5, + "trial_name": null, + "trial_params": null +}