diff --git "a/loss_log.txt" "b/loss_log.txt"
--- "a/loss_log.txt"
+++ "b/loss_log.txt"
@@ -1,647 +1,644 @@
-Step 1 | loss:1.221569538116455 lr:1.875e-06 tokens_per_second_per_gpu:776.4896475342558 
-Step 2 | loss:1.1539162397384644 lr:3.75e-06 tokens_per_second_per_gpu:2626.745768632657 
-Step 3 | loss:1.1532124280929565 lr:5.625e-06 tokens_per_second_per_gpu:3060.6355265050215 
-Step 4 | loss:0.9598733186721802 lr:7.5e-06 tokens_per_second_per_gpu:3063.150374667415 
-Step 5 | loss:1.020167589187622 lr:9.375000000000001e-06 tokens_per_second_per_gpu:2611.9293671879514 
-Step 6 | loss:1.1163296699523926 lr:1.125e-05 tokens_per_second_per_gpu:2225.818982541921 
-Step 7 | loss:1.077873706817627 lr:1.3125e-05 tokens_per_second_per_gpu:2603.7906186749615 
-Step 8 | loss:1.0882045030593872 lr:1.5e-05 tokens_per_second_per_gpu:2991.735662820424 
-Step 9 | loss:1.0246179103851318 lr:1.4999909358220728e-05 tokens_per_second_per_gpu:2410.161469741817 
-Step 10 | loss:0.8495532870292664 lr:1.4999637435073829e-05 tokens_per_second_per_gpu:3242.2852700467556 
-Step 11 | loss:0.9978005886077881 lr:1.4999184237131991e-05 tokens_per_second_per_gpu:3383.1990451125507 
-Step 12 | loss:0.9687494039535522 lr:1.499854977534953e-05 tokens_per_second_per_gpu:1902.8084244378674 
-Step 13 | loss:1.0104714632034302 lr:1.499773406506211e-05 tokens_per_second_per_gpu:2933.4949978737454 
-Step 14 | loss:0.7931515574455261 lr:1.4996737125986377e-05 tokens_per_second_per_gpu:4085.888567800566 
-Step 15 | loss:0.9825239777565002 lr:1.499555898221949e-05 tokens_per_second_per_gpu:2138.2084216687877 
-Step 16 | loss:0.8750104904174805 lr:1.4994199662238527e-05 tokens_per_second_per_gpu:3703.7749114369176 
-Step 17 | loss:0.8804910182952881 lr:1.4992659198899803e-05 tokens_per_second_per_gpu:2835.028146310613 
-Step 18 | loss:0.9359980821609497 lr:1.4990937629438073e-05 tokens_per_second_per_gpu:2644.538015822797 
-Step 19 | loss:0.8891004323959351 lr:1.4989034995465637e-05 tokens_per_second_per_gpu:2837.833590827904 
-Step 20 | loss:1.0624065399169922 lr:1.4986951342971331e-05 tokens_per_second_per_gpu:2155.890103956241 
-Step 21 | loss:0.9780190587043762 lr:1.498468672231941e-05 tokens_per_second_per_gpu:2565.782185800595 
-Step 22 | loss:1.071376919746399 lr:1.4982241188248342e-05 tokens_per_second_per_gpu:2007.8092025468436 
-Step 23 | loss:0.7347007989883423 lr:1.4979614799869475e-05 tokens_per_second_per_gpu:4224.400969539578 
-Step 24 | loss:0.9955159425735474 lr:1.4976807620665614e-05 tokens_per_second_per_gpu:2596.19245051982 
-Step 25 | loss:0.8271186947822571 lr:1.4973819718489483e-05 tokens_per_second_per_gpu:3168.5746698450243 
-Step 26 | loss:0.8975570201873779 lr:1.4970651165562086e-05 tokens_per_second_per_gpu:2739.229439199024 
-Step 27 | loss:1.089482307434082 lr:1.4967302038470966e-05 tokens_per_second_per_gpu:2347.5008041359756 
-Step 28 | loss:1.2102779150009155 lr:1.4963772418168344e-05 tokens_per_second_per_gpu:1994.4357047693568 
-Step 29 | loss:0.860817551612854 lr:1.4960062389969173e-05 tokens_per_second_per_gpu:2934.9487425459815 
-Step 30 | loss:0.9468755722045898 lr:1.4956172043549067e-05 tokens_per_second_per_gpu:2541.4215466420164 
-Step 31 | loss:1.0479042530059814 lr:1.4952101472942138e-05 tokens_per_second_per_gpu:2074.585452185979 
-Step 32 | loss:0.904244065284729 lr:1.4947850776538721e-05 tokens_per_second_per_gpu:2358.220020036641 
-Step 33 | loss:0.9157129526138306 lr:1.4943420057083003e-05 tokens_per_second_per_gpu:2490.175938191269 
-Step 34 | loss:0.8834834694862366 lr:1.4938809421670526e-05 tokens_per_second_per_gpu:3300.2406807038383 
-Step 35 | loss:0.9919298887252808 lr:1.493401898174561e-05 tokens_per_second_per_gpu:1849.696643954453 
-Step 36 | loss:0.9854268431663513 lr:1.4929048853098655e-05 tokens_per_second_per_gpu:2373.8255744270264 
-Step 37 | loss:0.8895089626312256 lr:1.492389915586334e-05 tokens_per_second_per_gpu:3564.264325106374 
-Step 38 | loss:1.0618623495101929 lr:1.491857001451373e-05 tokens_per_second_per_gpu:2546.3652101537164 
-Step 39 | loss:0.8307267427444458 lr:1.4913061557861243e-05 tokens_per_second_per_gpu:3101.1759165612048 
-Step 40 | loss:0.9268414974212646 lr:1.4907373919051571e-05 tokens_per_second_per_gpu:2604.8350849059198 
-Step 41 | loss:0.9643033146858215 lr:1.4901507235561428e-05 tokens_per_second_per_gpu:2546.930067535709 
-Step 42 | loss:0.8905935287475586 lr:1.4895461649195253e-05 tokens_per_second_per_gpu:3042.225207721806 
-Step 43 | loss:0.8909382224082947 lr:1.4889237306081767e-05 tokens_per_second_per_gpu:2220.37359285759 
-Step 44 | loss:1.0663670301437378 lr:1.4882834356670442e-05 tokens_per_second_per_gpu:1992.5821375012035 
-Step 45 | loss:0.7980119585990906 lr:1.4876252955727876e-05 tokens_per_second_per_gpu:2536.9415861871867 
-Step 46 | loss:0.9315576553344727 lr:1.486949326233404e-05 tokens_per_second_per_gpu:2015.857266504341 
-Step 47 | loss:0.818637490272522 lr:1.4862555439878434e-05 tokens_per_second_per_gpu:2764.9466270905377 
-Step 48 | loss:0.9148813486099243 lr:1.4855439656056144e-05 tokens_per_second_per_gpu:2458.8630962085804 
-Step 49 | loss:0.8426403403282166 lr:1.484814608286379e-05 tokens_per_second_per_gpu:3836.9937286372974 
-Step 50 | loss:0.9631166458129883 lr:1.4840674896595355e-05 tokens_per_second_per_gpu:2344.5407425233243 
-Step 51 | loss:0.9489616751670837 lr:1.4833026277837935e-05 tokens_per_second_per_gpu:2590.65407572909 
-Step 52 | loss:0.7524951696395874 lr:1.4825200411467377e-05 tokens_per_second_per_gpu:3316.8777964828005 
-Step 53 | loss:0.8084532618522644 lr:1.4817197486643802e-05 tokens_per_second_per_gpu:3023.881910604271 
-Step 54 | loss:0.8935003876686096 lr:1.480901769680703e-05 tokens_per_second_per_gpu:2796.6835981724653 
-Step 55 | loss:0.9641337990760803 lr:1.4800661239671925e-05 tokens_per_second_per_gpu:2619.6033697751586 
-Step 56 | loss:0.9214082360267639 lr:1.4792128317223583e-05 tokens_per_second_per_gpu:2223.910054733981 
-Step 57 | loss:0.8928937315940857 lr:1.4783419135712481e-05 tokens_per_second_per_gpu:2822.2559809661334 
-Step 58 | loss:0.9725896716117859 lr:1.4774533905649475e-05 tokens_per_second_per_gpu:3135.307257349326 
-Step 59 | loss:1.1065843105316162 lr:1.4765472841800714e-05 tokens_per_second_per_gpu:2383.5806852300307 
-Step 60 | loss:0.8419440984725952 lr:1.4756236163182447e-05 tokens_per_second_per_gpu:3695.211899955055 
-Step 61 | loss:0.8362283110618591 lr:1.4746824093055744e-05 tokens_per_second_per_gpu:3351.0990224903917 
-Step 62 | loss:0.9842981100082397 lr:1.4737236858921075e-05 tokens_per_second_per_gpu:3464.31229640519 
-Step 63 | loss:0.9068053364753723 lr:1.4727474692512831e-05 tokens_per_second_per_gpu:2966.1821865894576 
-Step 64 | loss:0.834957480430603 lr:1.4717537829793716e-05 tokens_per_second_per_gpu:2601.8903601590896 
-Step 65 | loss:0.7882446050643921 lr:1.4707426510949038e-05 tokens_per_second_per_gpu:3476.6321229392124 
-Step 66 | loss:0.799394965171814 lr:1.4697140980380914e-05 tokens_per_second_per_gpu:3120.9911279253056 
-Step 67 | loss:1.0695223808288574 lr:1.4686681486702354e-05 tokens_per_second_per_gpu:2008.6733315623778 
-Step 68 | loss:0.8479028940200806 lr:1.4676048282731259e-05 tokens_per_second_per_gpu:3461.3166586382476 
-Step 69 | loss:0.7161128520965576 lr:1.4665241625484297e-05 tokens_per_second_per_gpu:3323.2151026335396 
-Step 70 | loss:0.9997295141220093 lr:1.4654261776170711e-05 tokens_per_second_per_gpu:3499.3840144689457 
-Step 71 | loss:0.9570843577384949 lr:1.4643109000185986e-05 tokens_per_second_per_gpu:2803.3989631359914 
-Step 72 | loss:0.9544796347618103 lr:1.4631783567105444e-05 tokens_per_second_per_gpu:2907.0792063075314 
-Step 73 | loss:0.989665687084198 lr:1.4620285750677729e-05 tokens_per_second_per_gpu:2315.0531647541266 
-Step 74 | loss:0.8606690168380737 lr:1.460861582881818e-05 tokens_per_second_per_gpu:1761.639594790866 
-Step 75 | loss:0.8831307291984558 lr:1.4596774083602131e-05 tokens_per_second_per_gpu:2578.250829971227 
-Step 76 | loss:0.8336813449859619 lr:1.4584760801258072e-05 tokens_per_second_per_gpu:3966.441079179973 
-Step 77 | loss:1.0065804719924927 lr:1.4572576272160752e-05 tokens_per_second_per_gpu:2010.6956576896216 
-Step 78 | loss:0.9828115105628967 lr:1.4560220790824135e-05 tokens_per_second_per_gpu:2676.9438755928054 
-Step 79 | loss:0.8407449126243591 lr:1.4547694655894313e-05 tokens_per_second_per_gpu:3043.8163508680937 
-Step 80 | loss:0.8303096890449524 lr:1.4534998170142256e-05 tokens_per_second_per_gpu:2782.5148196337404 
-Step 81 | loss:0.7780717611312866 lr:1.4522131640456515e-05 tokens_per_second_per_gpu:3300.6096232500304 
-Step 82 | loss:0.7947230935096741 lr:1.4509095377835793e-05 tokens_per_second_per_gpu:3028.0607479145733 
-Step 83 | loss:0.8650649189949036 lr:1.4495889697381437e-05 tokens_per_second_per_gpu:3024.391338030527 
-Step 84 | loss:0.9109087586402893 lr:1.4482514918289811e-05 tokens_per_second_per_gpu:2882.55956475511 
-Step 85 | loss:0.8643304705619812 lr:1.4468971363844589e-05 tokens_per_second_per_gpu:2820.9174435328187 
-Step 86 | loss:0.8797863125801086 lr:1.4455259361408933e-05 tokens_per_second_per_gpu:3187.6221907265112 
-Step 87 | loss:1.0112887620925903 lr:1.4441379242417597e-05 tokens_per_second_per_gpu:2276.054803118254 
-Step 88 | loss:0.9944532513618469 lr:1.4427331342368889e-05 tokens_per_second_per_gpu:2373.6844209284745 
-Step 89 | loss:1.0218238830566406 lr:1.4413116000816587e-05 tokens_per_second_per_gpu:2183.503625916952 
-Step 90 | loss:0.9786317944526672 lr:1.439873356136172e-05 tokens_per_second_per_gpu:2308.142962546143 
-Step 91 | loss:1.0663024187088013 lr:1.4384184371644257e-05 tokens_per_second_per_gpu:2280.0466053008804 
-Step 92 | loss:0.9506876468658447 lr:1.436946878333472e-05 tokens_per_second_per_gpu:2214.0901274233033 
-Step 93 | loss:1.040751338005066 lr:1.435458715212567e-05 tokens_per_second_per_gpu:2776.228928077468 
-Step 94 | loss:0.8327547311782837 lr:1.4339539837723114e-05 tokens_per_second_per_gpu:2611.7891324599004 
-Step 95 | loss:0.8437411785125732 lr:1.4324327203837813e-05 tokens_per_second_per_gpu:2564.3074569788087 
-Step 96 | loss:0.8928688764572144 lr:1.4308949618176488e-05 tokens_per_second_per_gpu:3033.7427038894925 
-Step 97 | loss:0.8910201787948608 lr:1.4293407452432934e-05 tokens_per_second_per_gpu:2926.1872538932953 
-Step 98 | loss:1.1611385345458984 lr:1.4277701082279029e-05 tokens_per_second_per_gpu:2118.714336752679 
-Step 99 | loss:0.7676631212234497 lr:1.4261830887355666e-05 tokens_per_second_per_gpu:3119.234051386954 
-Step 100 | loss:0.9407196640968323 lr:1.4245797251263566e-05 tokens_per_second_per_gpu:3431.829631389003 
-Step 101 | loss:1.0271466970443726 lr:1.422960056155401e-05 tokens_per_second_per_gpu:2297.2551692634543 
-Step 102 | loss:0.7903147339820862 lr:1.4213241209719468e-05 tokens_per_second_per_gpu:3183.486695789974 
-Step 103 | loss:0.9327342510223389 lr:1.4196719591184145e-05 tokens_per_second_per_gpu:2308.249818645735 
-Step 104 | loss:1.1140373945236206 lr:1.4180036105294413e-05 tokens_per_second_per_gpu:2201.0614619912385 
-Step 105 | loss:0.8893833756446838 lr:1.4163191155309168e-05 tokens_per_second_per_gpu:2472.4719880913913 
-Step 106 | loss:0.8200833201408386 lr:1.4146185148390068e-05 tokens_per_second_per_gpu:3466.1216698989483 
-Step 107 | loss:0.9029530882835388 lr:1.4129018495591707e-05 tokens_per_second_per_gpu:2665.603306051977 
-Step 108 | loss:0.906825065612793 lr:1.4111691611851679e-05 tokens_per_second_per_gpu:2326.7020632435033 
-Step 109 | loss:0.852420449256897 lr:1.409420491598053e-05 tokens_per_second_per_gpu:2829.945359127956 
-Step 110 | loss:0.8402610421180725 lr:1.4076558830651655e-05 tokens_per_second_per_gpu:2424.5583345317846 
-Step 111 | loss:0.8761624097824097 lr:1.4058753782391077e-05 tokens_per_second_per_gpu:2714.916085371875 
-Step 112 | loss:1.0591604709625244 lr:1.4040790201567126e-05 tokens_per_second_per_gpu:1954.819773307979 
-Step 113 | loss:0.871858537197113 lr:1.4022668522380051e-05 tokens_per_second_per_gpu:1990.9860968133723 
-Step 114 | loss:0.9173691868782043 lr:1.4004389182851519e-05 tokens_per_second_per_gpu:3093.1846380992297 
-Step 115 | loss:0.7526230216026306 lr:1.3985952624814021e-05 tokens_per_second_per_gpu:3434.845480530211 
-Step 116 | loss:0.9190866947174072 lr:1.396735929390021e-05 tokens_per_second_per_gpu:2757.0267437124444 
-Step 117 | loss:0.9001188278198242 lr:1.3948609639532107e-05 tokens_per_second_per_gpu:3245.090811839929 
-Step 118 | loss:0.853203296661377 lr:1.3929704114910252e-05 tokens_per_second_per_gpu:2276.739974081572 
-Step 119 | loss:0.855078399181366 lr:1.3910643177002755e-05 tokens_per_second_per_gpu:2723.094778935103 
-Step 120 | loss:0.9485069513320923 lr:1.3891427286534234e-05 tokens_per_second_per_gpu:2207.578804777831 
-Step 121 | loss:0.8687245845794678 lr:1.3872056907974688e-05 tokens_per_second_per_gpu:2167.7910866884936 
-Step 122 | loss:0.7418152689933777 lr:1.3852532509528272e-05 tokens_per_second_per_gpu:3405.231105064688 
-Step 123 | loss:0.7311168909072876 lr:1.3832854563121978e-05 tokens_per_second_per_gpu:3483.8702435947193 
-Step 124 | loss:0.9829015731811523 lr:1.3813023544394226e-05 tokens_per_second_per_gpu:2585.438509633648 
-Step 125 | loss:0.8702593445777893 lr:1.3793039932683365e-05 tokens_per_second_per_gpu:2956.5308034715326 
-Step 126 | loss:0.927323579788208 lr:1.3772904211016098e-05 tokens_per_second_per_gpu:2484.745269269927 
-Step 127 | loss:0.947740375995636 lr:1.3752616866095793e-05 tokens_per_second_per_gpu:2078.82632153693 
-Step 128 | loss:0.9461579918861389 lr:1.373217838829073e-05 tokens_per_second_per_gpu:2705.8367865703003 
-Step 129 | loss:0.8957231044769287 lr:1.3711589271622237e-05 tokens_per_second_per_gpu:2992.601213940647 
-Step 130 | loss:0.8435050249099731 lr:1.369085001375276e-05 tokens_per_second_per_gpu:2422.468716757249 
-Step 131 | loss:0.8020814657211304 lr:1.3669961115973831e-05 tokens_per_second_per_gpu:2716.525699767116 
-Step 132 | loss:1.0148298740386963 lr:1.3648923083193945e-05 tokens_per_second_per_gpu:3222.4037213065576 
-Step 133 | loss:0.8540488481521606 lr:1.3627736423926358e-05 tokens_per_second_per_gpu:3156.5818359892814 
-Step 134 | loss:1.0741864442825317 lr:1.360640165027681e-05 tokens_per_second_per_gpu:3006.5484287697604 
-Step 135 | loss:0.9130982756614685 lr:1.3584919277931113e-05 tokens_per_second_per_gpu:3060.7231378677075 
-Step 136 | loss:0.9075719118118286 lr:1.3563289826142734e-05 tokens_per_second_per_gpu:2994.295564077511 
-Step 137 | loss:0.7380925416946411 lr:1.3541513817720196e-05 tokens_per_second_per_gpu:3380.35958901387 
-Step 138 | loss:0.7887682914733887 lr:1.351959177901448e-05 tokens_per_second_per_gpu:3411.1643530753004 
-Step 139 | loss:1.0572377443313599 lr:1.3497524239906268e-05 tokens_per_second_per_gpu:1846.5574752328034 
-Step 140 | loss:0.9169989824295044 lr:1.3475311733793171e-05 tokens_per_second_per_gpu:2659.9050056281694 
-Step 141 | loss:0.7694222331047058 lr:1.3452954797576803e-05 tokens_per_second_per_gpu:2248.619441667407 
-Step 142 | loss:0.9348461627960205 lr:1.343045397164983e-05 tokens_per_second_per_gpu:2748.0654635407595 
-Step 143 | loss:1.0059735774993896 lr:1.3407809799882887e-05 tokens_per_second_per_gpu:2571.4433195970273 
-Step 144 | loss:0.9550279974937439 lr:1.338502282961145e-05 tokens_per_second_per_gpu:2268.0708636948866 
-Step 145 | loss:0.975806713104248 lr:1.3362093611622587e-05 tokens_per_second_per_gpu:2522.5177215146223 
-Step 146 | loss:0.7611340284347534 lr:1.333902270014167e-05 tokens_per_second_per_gpu:2906.4275538550883 
-Step 147 | loss:0.9788836240768433 lr:1.3315810652818951e-05 tokens_per_second_per_gpu:3477.341552657733 
-Step 148 | loss:0.8796938061714172 lr:1.3292458030716105e-05 tokens_per_second_per_gpu:2631.896237909145 
-Step 149 | loss:0.9417287707328796 lr:1.3268965398292655e-05 tokens_per_second_per_gpu:3192.0815935061937 
-Step 150 | loss:1.0355844497680664 lr:1.3245333323392335e-05 tokens_per_second_per_gpu:2245.145788317112 
-Step 151 | loss:0.8991612195968628 lr:1.3221562377229365e-05 tokens_per_second_per_gpu:2578.650347755809 
-Step 152 | loss:0.9595103859901428 lr:1.3197653134374637e-05 tokens_per_second_per_gpu:2434.651933495438 
-Step 153 | loss:0.9001013040542603 lr:1.3173606172741836e-05 tokens_per_second_per_gpu:1842.5350349006453 
-Step 154 | loss:0.8173960447311401 lr:1.3149422073573466e-05 tokens_per_second_per_gpu:2488.3925420974924 
-Step 155 | loss:0.8045638203620911 lr:1.3125101421426801e-05 tokens_per_second_per_gpu:3302.756369196822 
-Step 156 | loss:0.7967942357063293 lr:1.3100644804159756e-05 tokens_per_second_per_gpu:2950.507653722916 
-Step 157 | loss:0.9466549754142761 lr:1.307605281291668e-05 tokens_per_second_per_gpu:2554.2876959141595 
-Step 158 | loss:0.7874048352241516 lr:1.3051326042114062e-05 tokens_per_second_per_gpu:2503.830061713888 
-Step 159 | loss:0.7649244070053101 lr:1.3026465089426172e-05 tokens_per_second_per_gpu:3764.466272015238 
-Step 160 | loss:0.9769793152809143 lr:1.3001470555770603e-05 tokens_per_second_per_gpu:2134.8838608368396 
-Step 161 | loss:0.6409870386123657 lr:1.2976343045293758e-05 tokens_per_second_per_gpu:3831.6710466709137 
-Step 162 | loss:0.9095283150672913 lr:1.2951083165356238e-05 tokens_per_second_per_gpu:2975.461143498561 
-Step 163 | loss:0.987684428691864 lr:1.2925691526518168e-05 tokens_per_second_per_gpu:2419.6660438127183 
-Step 164 | loss:0.8870441317558289 lr:1.2900168742524431e-05 tokens_per_second_per_gpu:2720.042439707338 
-Step 165 | loss:1.0281869173049927 lr:1.2874515430289843e-05 tokens_per_second_per_gpu:2667.0587877368976 
-Step 166 | loss:0.9013745784759521 lr:1.2848732209884236e-05 tokens_per_second_per_gpu:2421.311690375509 
-Step 167 | loss:0.8012135028839111 lr:1.2822819704517467e-05 tokens_per_second_per_gpu:2909.767624354188 
-Step 168 | loss:0.8641454577445984 lr:1.2796778540524362e-05 tokens_per_second_per_gpu:3065.6247029851916 
-Step 169 | loss:0.8661941885948181 lr:1.277060934734957e-05 tokens_per_second_per_gpu:2696.294949936513 
-Step 170 | loss:0.9361519813537598 lr:1.2744312757532357e-05 tokens_per_second_per_gpu:3128.4635750227962 
-Step 171 | loss:0.9445534944534302 lr:1.2717889406691301e-05 tokens_per_second_per_gpu:2804.1901845669454 
-Step 172 | loss:0.8409781455993652 lr:1.2691339933508946e-05 tokens_per_second_per_gpu:2300.349952366563 
-Step 173 | loss:0.7127365469932556 lr:1.266466497971636e-05 tokens_per_second_per_gpu:3292.686682785377 
-Step 174 | loss:0.9441195726394653 lr:1.263786519007761e-05 tokens_per_second_per_gpu:2457.8393080536093 
-Step 175 | loss:0.8693594336509705 lr:1.2610941212374196e-05 tokens_per_second_per_gpu:2543.008326528785 
-Step 176 | loss:0.8825166821479797 lr:1.2583893697389384e-05 tokens_per_second_per_gpu:2725.6009677627276 
-Step 177 | loss:0.8768410682678223 lr:1.2556723298892478e-05 tokens_per_second_per_gpu:3557.1565324485255 
-Step 178 | loss:1.1754149198532104 lr:1.252943067362301e-05 tokens_per_second_per_gpu:2417.834316522929 
-Step 179 | loss:0.8500137329101562 lr:1.250201648127488e-05 tokens_per_second_per_gpu:2250.918842990208 
-Step 180 | loss:1.1264166831970215 lr:1.2474481384480403e-05 tokens_per_second_per_gpu:2391.435699419417 
-Step 181 | loss:0.7425944805145264 lr:1.2446826048794285e-05 tokens_per_second_per_gpu:3704.124439704411 
-Step 182 | loss:0.9915411472320557 lr:1.2419051142677552e-05 tokens_per_second_per_gpu:2500.2454386416653 
-Step 183 | loss:1.094246506690979 lr:1.2391157337481379e-05 tokens_per_second_per_gpu:2292.6884961819756 
-Step 184 | loss:0.8413046002388 lr:1.2363145307430866e-05 tokens_per_second_per_gpu:3114.0205371534985 
-Step 185 | loss:0.8624469041824341 lr:1.2335015729608747e-05 tokens_per_second_per_gpu:2642.208561896737 
-Step 186 | loss:0.8683147430419922 lr:1.2306769283939017e-05 tokens_per_second_per_gpu:3871.095670001607 
-Step 187 | loss:0.9512786865234375 lr:1.2278406653170503e-05 tokens_per_second_per_gpu:2628.580728112808 
-Step 188 | loss:0.8004306554794312 lr:1.2249928522860353e-05 tokens_per_second_per_gpu:3667.905766206227 
-Step 189 | loss:0.8889120221138 lr:1.2221335581357475e-05 tokens_per_second_per_gpu:2248.7450084836746 
-Step 190 | loss:0.8228364586830139 lr:1.2192628519785897e-05 tokens_per_second_per_gpu:2514.6297054243464 
-Step 191 | loss:0.944223165512085 lr:1.2163808032028051e-05 tokens_per_second_per_gpu:2934.22518036366 
-Step 192 | loss:0.7095787525177002 lr:1.213487481470802e-05 tokens_per_second_per_gpu:3819.992979162668 
-Step 193 | loss:0.8380393981933594 lr:1.2105829567174679e-05 tokens_per_second_per_gpu:2800.2528555689605 
-Step 194 | loss:1.0584007501602173 lr:1.2076672991484807e-05 tokens_per_second_per_gpu:2438.3413375541963 
-Step 195 | loss:1.125288486480713 lr:1.2047405792386106e-05 tokens_per_second_per_gpu:1937.2355566529484 
-Step 196 | loss:0.9059362411499023 lr:1.2018028677300183e-05 tokens_per_second_per_gpu:2153.221698146553 
-Step 197 | loss:0.8372734189033508 lr:1.1988542356305431e-05 tokens_per_second_per_gpu:2404.6834628720703 
-Step 198 | loss:0.8674279451370239 lr:1.1958947542119874e-05 tokens_per_second_per_gpu:2321.484011546196 
-Step 199 | loss:0.8657681941986084 lr:1.1929244950083946e-05 tokens_per_second_per_gpu:2348.2580286455236 
-Step 200 | loss:0.9470924139022827 lr:1.1899435298143186e-05 tokens_per_second_per_gpu:2393.2309464100285 
-Step 201 | loss:0.892507791519165 lr:1.1869519306830906e-05 tokens_per_second_per_gpu:3206.4545463886843 
-Step 202 | loss:0.9081307649612427 lr:1.1839497699250747e-05 tokens_per_second_per_gpu:3404.974687265603 
-Step 203 | loss:0.8427631855010986 lr:1.1809371201059226e-05 tokens_per_second_per_gpu:3047.938780516295 
-Step 204 | loss:0.7953866720199585 lr:1.1779140540448186e-05 tokens_per_second_per_gpu:2537.5554252587053 
-Step 205 | loss:0.8499541878700256 lr:1.1748806448127187e-05 tokens_per_second_per_gpu:3022.9311446708475 
-Step 206 | loss:0.9877169132232666 lr:1.1718369657305856e-05 tokens_per_second_per_gpu:1923.481770749864 
-Step 207 | loss:0.9637613892555237 lr:1.1687830903676162e-05 tokens_per_second_per_gpu:1922.9230396108537 
-Step 208 | loss:0.8312227129936218 lr:1.1657190925394627e-05 tokens_per_second_per_gpu:2981.907027352506 
-Step 209 | loss:0.7979338765144348 lr:1.1626450463064492e-05 tokens_per_second_per_gpu:2995.4292103615694 
-Step 210 | loss:0.8698200583457947 lr:1.1595610259717812e-05 tokens_per_second_per_gpu:2902.551497227684 
-Step 211 | loss:0.8625684380531311 lr:1.1564671060797493e-05 tokens_per_second_per_gpu:2660.138583168492 
-Step 212 | loss:0.7270165681838989 lr:1.1533633614139274e-05 tokens_per_second_per_gpu:2835.768233651723 
-Step 213 | loss:0.9477829337120056 lr:1.1502498669953669e-05 tokens_per_second_per_gpu:2406.7431958408943 
-Step 214 | loss:0.7979986071586609 lr:1.1471266980807803e-05 tokens_per_second_per_gpu:2528.5572177691274 
-Step 215 | loss:0.8460088968276978 lr:1.143993930160724e-05 tokens_per_second_per_gpu:2736.724211562585 
-Step 216 | loss:0.8999170660972595 lr:1.1408516389577742e-05 tokens_per_second_per_gpu:2756.4852763216604 
-Step 217 | loss:0.9149824380874634 lr:1.1376999004246942e-05 tokens_per_second_per_gpu:2998.14918902869 
-Step 218 | loss:1.035408616065979 lr:1.1345387907426016e-05 tokens_per_second_per_gpu:2247.566429188033 
-Step 219 | loss:0.9593961238861084 lr:1.1313683863191246e-05 tokens_per_second_per_gpu:2672.649906422322 
-Step 220 | loss:0.9525635242462158 lr:1.1281887637865556e-05 tokens_per_second_per_gpu:2702.378081627573 
-Step 221 | loss:0.8132772445678711 lr:1.125e-05 tokens_per_second_per_gpu:3544.073171959475 
-Step 222 | loss:0.8736097812652588 lr:1.1218021720355171e-05 tokens_per_second_per_gpu:3763.371300035145 
-Step 223 | loss:0.8046478033065796 lr:1.1185953571882582e-05 tokens_per_second_per_gpu:2600.942499066943 
-Step 224 | loss:0.8822525143623352 lr:1.1153796329705974e-05 tokens_per_second_per_gpu:3243.5748920774936 
-Step 225 | loss:1.209902048110962 lr:1.1121550771102586e-05 tokens_per_second_per_gpu:1669.9666501667505 
-Step 226 | loss:0.832476794719696 lr:1.108921767548437e-05 tokens_per_second_per_gpu:2864.4252757404315 
-Step 227 | loss:0.8101415634155273 lr:1.1056797824379137e-05 tokens_per_second_per_gpu:2598.617402658873 
-Step 228 | loss:0.9285197854042053 lr:1.1024292001411689e-05 tokens_per_second_per_gpu:2088.9722996943365 
-Step 229 | loss:0.8414878845214844 lr:1.0991700992284858e-05 tokens_per_second_per_gpu:3063.9756544789043 
-Step 230 | loss:0.9940211772918701 lr:1.0959025584760526e-05 tokens_per_second_per_gpu:2433.3544485786088 
-Step 231 | loss:0.8977429270744324 lr:1.0926266568640585e-05 tokens_per_second_per_gpu:2542.186006721706 
-Step 232 | loss:0.9585859179496765 lr:1.089342473574783e-05 tokens_per_second_per_gpu:2300.665172041385 
-Step 233 | loss:0.9505215883255005 lr:1.0860500879906838e-05 tokens_per_second_per_gpu:3180.361033842941 
-Step 234 | loss:0.8571861982345581 lr:1.0827495796924784e-05 tokens_per_second_per_gpu:2594.923533596223 
-Step 235 | loss:0.7735648155212402 lr:1.0794410284572181e-05 tokens_per_second_per_gpu:2694.024955898258 
-Step 236 | loss:0.8936963677406311 lr:1.0761245142563622e-05 tokens_per_second_per_gpu:2670.446505293131 
-Step 237 | loss:0.9809367656707764 lr:1.0728001172538428e-05 tokens_per_second_per_gpu:2571.567322130418 
-Step 238 | loss:0.9553939700126648 lr:1.06946791780413e-05 tokens_per_second_per_gpu:2541.26905948664 
-Step 239 | loss:0.7851515412330627 lr:1.0661279964502861e-05 tokens_per_second_per_gpu:3012.100826567419 
-Step 240 | loss:0.8377625346183777 lr:1.0627804339220218e-05 tokens_per_second_per_gpu:2908.6314595373565 
-Step 241 | loss:1.0584880113601685 lr:1.0594253111337442e-05 tokens_per_second_per_gpu:2127.0916213056576 
-Step 242 | loss:0.8319791555404663 lr:1.0560627091825987e-05 tokens_per_second_per_gpu:2945.196078622988 
-Step 243 | loss:0.9787124395370483 lr:1.0526927093465123e-05 tokens_per_second_per_gpu:2276.353220549065 
-Step 244 | loss:0.9965022206306458 lr:1.0493153930822264e-05 tokens_per_second_per_gpu:2830.526282612664 
-Step 245 | loss:0.8443072438240051 lr:1.0459308420233291e-05 tokens_per_second_per_gpu:2699.7709016776953 
-Step 246 | loss:0.8488413691520691 lr:1.042539137978282e-05 tokens_per_second_per_gpu:2581.172056006769 
-Step 247 | loss:0.7844489812850952 lr:1.039140362928442e-05 tokens_per_second_per_gpu:3944.3421615811703 
-Step 248 | loss:0.8647257685661316 lr:1.0357345990260806e-05 tokens_per_second_per_gpu:2875.752228028819 
-Step 249 | loss:0.8394876718521118 lr:1.0323219285923976e-05 tokens_per_second_per_gpu:2754.3186309441694 
-Step 250 | loss:0.8988162279129028 lr:1.0289024341155321e-05 tokens_per_second_per_gpu:2413.305312945081 
-Step 251 | loss:0.8642686605453491 lr:1.0254761982485678e-05 tokens_per_second_per_gpu:2703.537161585324 
-Step 252 | loss:0.9334269165992737 lr:1.0220433038075347e-05 tokens_per_second_per_gpu:2806.6086135918135 
-Step 253 | loss:0.8606577515602112 lr:1.0186038337694097e-05 tokens_per_second_per_gpu:2776.029742452688 
-Step 254 | loss:0.9714118838310242 lr:1.0151578712701077e-05 tokens_per_second_per_gpu:2382.253753459991 
-Step 255 | loss:0.865350067615509 lr:1.0117054996024753e-05 tokens_per_second_per_gpu:2716.911389419169 
-Step 256 | loss:0.8225361704826355 lr:1.008246802214275e-05 tokens_per_second_per_gpu:2850.587447893 
-Step 257 | loss:0.7821142673492432 lr:1.0047818627061696e-05 tokens_per_second_per_gpu:3493.851190905611 
-Step 258 | loss:0.9727859497070312 lr:1.0013107648297013e-05 tokens_per_second_per_gpu:2163.1960733975143 
-Step 259 | loss:0.8025305271148682 lr:9.978335924852662e-06 tokens_per_second_per_gpu:2213.270287504602 
-Step 260 | loss:0.7221256494522095 lr:9.943504297200883e-06 tokens_per_second_per_gpu:4165.593745585308 
-Step 261 | loss:1.0640528202056885 lr:9.908613607261861e-06 tokens_per_second_per_gpu:2175.662315838582 
-Step 262 | loss:0.8858134746551514 lr:9.873664698383386e-06 tokens_per_second_per_gpu:2108.5291412185948 
-Step 263 | loss:0.997913122177124 lr:9.838658415320474e-06 tokens_per_second_per_gpu:2297.180689912119 
-Step 264 | loss:0.8774775266647339 lr:9.803595604214924e-06 tokens_per_second_per_gpu:2114.8571076658072 
-Step 265 | loss:0.8805571794509888 lr:9.768477112574902e-06 tokens_per_second_per_gpu:3400.560855606231 
-Step 266 | loss:0.9601130485534668 lr:9.733303789254418e-06 tokens_per_second_per_gpu:2678.9658431685334 
-Step 267 | loss:0.9212712049484253 lr:9.698076484432837e-06 tokens_per_second_per_gpu:2042.823107268897 
-Step 268 | loss:0.7927682399749756 lr:9.662796049594319e-06 tokens_per_second_per_gpu:3287.43519927435 
-Step 269 | loss:1.1724016666412354 lr:9.627463337507226e-06 tokens_per_second_per_gpu:2966.2408737525116 
-Step 270 | loss:0.9678674936294556 lr:9.592079202203536e-06 tokens_per_second_per_gpu:2884.935388288082 
-Step 271 | loss:0.8864798545837402 lr:9.556644498958176e-06 tokens_per_second_per_gpu:2400.49316338112 
-Step 272 | loss:0.938960075378418 lr:9.521160084268355e-06 tokens_per_second_per_gpu:2212.7791306773547 
-Step 273 | loss:0.8877987265586853 lr:9.485626815832877e-06 tokens_per_second_per_gpu:3354.291244394353 
-Step 274 | loss:0.8895339369773865 lr:9.450045552531383e-06 tokens_per_second_per_gpu:3240.0340597607606 
-Step 275 | loss:0.8576871752738953 lr:9.414417154403609e-06 tokens_per_second_per_gpu:3453.885492750505 
-Step 276 | loss:0.9489049315452576 lr:9.378742482628598e-06 tokens_per_second_per_gpu:3090.6101442457916 
-Step 277 | loss:0.899044394493103 lr:9.343022399503873e-06 tokens_per_second_per_gpu:2594.79916468504 
-Step 278 | loss:0.8113265633583069 lr:9.307257768424607e-06 tokens_per_second_per_gpu:3177.796880529129 
-Step 279 | loss:0.9940325617790222 lr:9.271449453862748e-06 tokens_per_second_per_gpu:2404.427909840856 
-Step 280 | loss:0.8677517771720886 lr:9.235598321346119e-06 tokens_per_second_per_gpu:3642.3763622655156 
-Step 281 | loss:0.9939593076705933 lr:9.1997052374375e-06 tokens_per_second_per_gpu:2429.8607833688475 
-Step 282 | loss:0.9184926748275757 lr:9.163771069713696e-06 tokens_per_second_per_gpu:2404.497162239973 
-Step 283 | loss:0.8467150926589966 lr:9.12779668674454e-06 tokens_per_second_per_gpu:3138.575898512493 
-Step 284 | loss:0.892140805721283 lr:9.091782958071923e-06 tokens_per_second_per_gpu:2558.36640338516 
-Step 285 | loss:0.747302234172821 lr:9.055730754188765e-06 tokens_per_second_per_gpu:2942.3912091461602 
-Step 286 | loss:0.857389509677887 lr:9.019640946517972e-06 tokens_per_second_per_gpu:2200.4241563666415 
-Step 287 | loss:0.9071866869926453 lr:8.98351440739138e-06 tokens_per_second_per_gpu:2901.754213786485 
-Step 288 | loss:0.9092150926589966 lr:8.947352010028666e-06 tokens_per_second_per_gpu:2387.9788412693297 
-Step 289 | loss:0.9917126297950745 lr:8.911154628516236e-06 tokens_per_second_per_gpu:3061.2987875670756 
-Step 290 | loss:0.9317039847373962 lr:8.874923137786114e-06 tokens_per_second_per_gpu:3225.279467510707 
-Step 291 | loss:0.7867501974105835 lr:8.838658413594773e-06 tokens_per_second_per_gpu:3152.63949537652 
-Step 292 | loss:0.9737382531166077 lr:8.80236133250198e-06 tokens_per_second_per_gpu:2712.0793822195424 
-Step 293 | loss:0.8233653903007507 lr:8.766032771849601e-06 tokens_per_second_per_gpu:2616.28775170671 
-Step 294 | loss:0.8503261208534241 lr:8.729673609740411e-06 tokens_per_second_per_gpu:3122.336820825079 
-Step 295 | loss:0.831436038017273 lr:8.693284725016845e-06 tokens_per_second_per_gpu:2756.8460814254904 
-Step 296 | loss:0.8883024454116821 lr:8.656866997239767e-06 tokens_per_second_per_gpu:2849.1394403891695 
-Step 297 | loss:1.030733346939087 lr:8.620421306667225e-06 tokens_per_second_per_gpu:2177.6471527849394 
-Step 298 | loss:0.880768358707428 lr:8.58394853423314e-06 tokens_per_second_per_gpu:2537.216693346244 
-Step 299 | loss:0.8131841421127319 lr:8.547449561526045e-06 tokens_per_second_per_gpu:3331.622489013824 
-Step 300 | loss:0.904315710067749 lr:8.510925270767766e-06 tokens_per_second_per_gpu:2341.584545888117 
-Step 301 | loss:0.8966640830039978 lr:8.474376544792087e-06 tokens_per_second_per_gpu:2190.578843299298 
-Step 302 | loss:0.8454634547233582 lr:8.43780426702342e-06 tokens_per_second_per_gpu:2610.455821912868 
-Step 303 | loss:0.8394374847412109 lr:8.40120932145545e-06 tokens_per_second_per_gpu:3411.2844261560285 
-Step 304 | loss:0.8339081406593323 lr:8.36459259262978e-06 tokens_per_second_per_gpu:2536.6041667350532 
-Step 305 | loss:0.87993323802948 lr:8.327954965614526e-06 tokens_per_second_per_gpu:2955.8406787717813 
-Step 306 | loss:0.8896875977516174 lr:8.291297325982935e-06 tokens_per_second_per_gpu:2856.849362787982 
-Step 307 | loss:0.8964165449142456 lr:8.254620559792e-06 tokens_per_second_per_gpu:2514.582769533579 
-Step 308 | loss:0.8504213094711304 lr:8.217925553561006e-06 tokens_per_second_per_gpu:2586.2976029121282 
-Step 309 | loss:0.9949404001235962 lr:8.18121319425013e-06 tokens_per_second_per_gpu:2718.513680534231 
-Step 310 | loss:0.8377817273139954 lr:8.14448436923899e-06 tokens_per_second_per_gpu:2479.359962245908 
-Step 311 | loss:0.9631475806236267 lr:8.1077399663052e-06 tokens_per_second_per_gpu:2560.2062470472647 
-Step 312 | loss:0.8673937320709229 lr:8.070980873602909e-06 tokens_per_second_per_gpu:2299.3708471618284 
-Step 313 | loss:0.7618104815483093 lr:8.034207979641328e-06 tokens_per_second_per_gpu:2870.2023552491078 
-Step 314 | loss:0.9050427675247192 lr:7.99742217326328e-06 tokens_per_second_per_gpu:2367.7275558955394 
-Step 315 | loss:0.8086662888526917 lr:7.960624343623674e-06 tokens_per_second_per_gpu:2866.746041665019 
-Step 316 | loss:0.8907902836799622 lr:7.923815380168046e-06 tokens_per_second_per_gpu:2496.810028140203 
-Step 317 | loss:0.9406841993331909 lr:7.886996172611049e-06 tokens_per_second_per_gpu:2524.8161688710366 
-Step 318 | loss:0.8578323125839233 lr:7.850167610914942e-06 tokens_per_second_per_gpu:3224.289434866302 
-Step 319 | loss:0.886796236038208 lr:7.813330585268092e-06 tokens_per_second_per_gpu:3023.1847564125983 
-Step 320 | loss:0.8145670890808105 lr:7.776485986063437e-06 tokens_per_second_per_gpu:2596.918537084552 
-Step 321 | loss:0.9760152697563171 lr:7.73963470387699e-06 tokens_per_second_per_gpu:2434.8368875247716 
-Step 322 | loss:0.8733581304550171 lr:7.702777629446298e-06 tokens_per_second_per_gpu:2350.605592846224 
-Step 323 | loss:0.9446771144866943 lr:7.665915653648901e-06 tokens_per_second_per_gpu:2151.967467013007 
-Step 324 | loss:1.06463623046875 lr:7.629049667480826e-06 tokens_per_second_per_gpu:2519.802402001832 
-Step 325 | loss:0.8435251116752625 lr:7.592180562035022e-06 tokens_per_second_per_gpu:3336.522359641145 
-Step 326 | loss:0.8608092069625854 lr:7.555309228479843e-06 tokens_per_second_per_gpu:3277.118785268545 
-Step 327 | loss:0.7815330028533936 lr:7.518436558037498e-06 tokens_per_second_per_gpu:3369.0019186297413 
-Step 328 | loss:0.8153398036956787 lr:7.481563441962503e-06 tokens_per_second_per_gpu:3267.182160183334 
-Step 329 | loss:0.8034992814064026 lr:7.444690771520157e-06 tokens_per_second_per_gpu:3690.7640838779034 
-Step 330 | loss:0.8502854108810425 lr:7.407819437964981e-06 tokens_per_second_per_gpu:2631.1245605629197 
-Step 331 | loss:0.8993874788284302 lr:7.370950332519176e-06 tokens_per_second_per_gpu:2930.8409402558013 
-Step 332 | loss:0.9783048629760742 lr:7.3340843463511e-06 tokens_per_second_per_gpu:2899.8105315950947 
-Step 333 | loss:0.9575924277305603 lr:7.2972223705537036e-06 tokens_per_second_per_gpu:2429.0229008539395 
-Step 334 | loss:0.865286648273468 lr:7.26036529612301e-06 tokens_per_second_per_gpu:2839.6573926637784 
-Step 335 | loss:0.8176185488700867 lr:7.223514013936566e-06 tokens_per_second_per_gpu:3094.1096527236145 
-Step 336 | loss:0.7247341275215149 lr:7.186669414731913e-06 tokens_per_second_per_gpu:3315.033496164722 
-Step 337 | loss:0.8403685092926025 lr:7.14983238908506e-06 tokens_per_second_per_gpu:3039.9672711448557 
-Step 338 | loss:0.7528369426727295 lr:7.1130038273889515e-06 tokens_per_second_per_gpu:3136.680127173355 
-Step 339 | loss:0.8461508750915527 lr:7.0761846198319535e-06 tokens_per_second_per_gpu:2716.7048475070237 
-Step 340 | loss:1.0012428760528564 lr:7.039375656376329e-06 tokens_per_second_per_gpu:2181.341224281936 
-Step 341 | loss:0.8235898017883301 lr:7.0025778267367225e-06 tokens_per_second_per_gpu:3569.9367239582057 
-Step 342 | loss:1.0386333465576172 lr:6.965792020358672e-06 tokens_per_second_per_gpu:2647.7206930607267 
-Step 343 | loss:0.900693953037262 lr:6.9290191263970935e-06 tokens_per_second_per_gpu:2909.1159397211577 
-Step 344 | loss:0.8273099660873413 lr:6.8922600336948e-06 tokens_per_second_per_gpu:2416.7004437081237 
-Step 345 | loss:0.9471863508224487 lr:6.8555156307610125e-06 tokens_per_second_per_gpu:2800.786061572154 
-Step 346 | loss:1.011666178703308 lr:6.818786805749872e-06 tokens_per_second_per_gpu:3052.515712984993 
-Step 347 | loss:0.8267448544502258 lr:6.782074446438995e-06 tokens_per_second_per_gpu:2917.6837742563043 
-Step 348 | loss:0.8594385385513306 lr:6.745379440208001e-06 tokens_per_second_per_gpu:3103.4111269442988 
-Step 349 | loss:1.0058006048202515 lr:6.708702674017064e-06 tokens_per_second_per_gpu:2278.381957741059 
-Step 350 | loss:0.9441714286804199 lr:6.672045034385478e-06 tokens_per_second_per_gpu:2718.044038121958 
-Step 351 | loss:0.9618679285049438 lr:6.635407407370222e-06 tokens_per_second_per_gpu:3258.8267915094248 
-Step 352 | loss:0.9324498176574707 lr:6.598790678544549e-06 tokens_per_second_per_gpu:2416.68747120486 
-Step 353 | loss:0.8107538819313049 lr:6.562195732976582e-06 tokens_per_second_per_gpu:2136.3859791339296 
-Step 354 | loss:0.8707559108734131 lr:6.525623455207914e-06 tokens_per_second_per_gpu:3920.9124449450605 
-Step 355 | loss:1.0396661758422852 lr:6.489074729232236e-06 tokens_per_second_per_gpu:2677.917087444475 
-Step 356 | loss:0.9951695203781128 lr:6.452550438473955e-06 tokens_per_second_per_gpu:2513.724453514746 
-Step 357 | loss:0.8180606365203857 lr:6.416051465766861e-06 tokens_per_second_per_gpu:2136.395063021162 
-Step 358 | loss:0.8964337706565857 lr:6.379578693332777e-06 tokens_per_second_per_gpu:2502.99951437964 
-Step 359 | loss:0.8790155053138733 lr:6.343133002760231e-06 tokens_per_second_per_gpu:2575.228954991082 
-Step 360 | loss:0.8716058731079102 lr:6.30671527498316e-06 tokens_per_second_per_gpu:2902.1444594976692 
-Step 361 | loss:0.8309515714645386 lr:6.270326390259591e-06 tokens_per_second_per_gpu:2755.057233486617 
-Step 362 | loss:1.0263001918792725 lr:6.233967228150399e-06 tokens_per_second_per_gpu:2805.8836613470926 
-Step 363 | loss:0.951583206653595 lr:6.197638667498023e-06 tokens_per_second_per_gpu:2874.3500077114672 
-Step 364 | loss:0.7161557078361511 lr:6.161341586405229e-06 tokens_per_second_per_gpu:3021.893258538902 
-Step 365 | loss:0.9259535074234009 lr:6.125076862213888e-06 tokens_per_second_per_gpu:2695.4417612817692 
-Step 366 | loss:1.016567587852478 lr:6.088845371483765e-06 tokens_per_second_per_gpu:3001.6669903895245 
-Step 367 | loss:0.9344175457954407 lr:6.0526479899713365e-06 tokens_per_second_per_gpu:2613.816790424833 
-Step 368 | loss:0.9176039695739746 lr:6.0164855926086204e-06 tokens_per_second_per_gpu:2374.669316445514 
-Step 369 | loss:0.8734990358352661 lr:5.980359053482029e-06 tokens_per_second_per_gpu:2290.221515119589 
-Step 370 | loss:0.8557796478271484 lr:5.944269245811238e-06 tokens_per_second_per_gpu:2167.941273792827 
-Step 371 | loss:0.8427585363388062 lr:5.908217041928078e-06 tokens_per_second_per_gpu:2712.4728476039068 
-Step 372 | loss:0.7605788111686707 lr:5.87220331325546e-06 tokens_per_second_per_gpu:2384.5312207906018 
-Step 373 | loss:0.8790305256843567 lr:5.836228930286305e-06 tokens_per_second_per_gpu:2610.707228443902 
-Step 374 | loss:1.0703303813934326 lr:5.800294762562498e-06 tokens_per_second_per_gpu:1830.7243589924585 
-Step 375 | loss:0.8680859804153442 lr:5.764401678653884e-06 tokens_per_second_per_gpu:3279.2616810969766 
-Step 376 | loss:0.822643518447876 lr:5.728550546137254e-06 tokens_per_second_per_gpu:2555.8667929702524 
-Step 377 | loss:0.8699767589569092 lr:5.692742231575392e-06 tokens_per_second_per_gpu:3389.3087628096314 
-Step 378 | loss:0.8780330419540405 lr:5.656977600496128e-06 tokens_per_second_per_gpu:2654.4637897137886 
-Step 379 | loss:0.7730309367179871 lr:5.621257517371404e-06 tokens_per_second_per_gpu:3445.66256348602 
-Step 380 | loss:0.8410382270812988 lr:5.585582845596393e-06 tokens_per_second_per_gpu:3472.3318332317604 
-Step 381 | loss:0.9238551259040833 lr:5.54995444746862e-06 tokens_per_second_per_gpu:2329.4743010850307 
-Step 382 | loss:1.0120571851730347 lr:5.514373184167124e-06 tokens_per_second_per_gpu:2835.2976676586004 
-Step 383 | loss:0.8865228891372681 lr:5.478839915731643e-06 tokens_per_second_per_gpu:2756.7384007998317 
-Step 384 | loss:0.7867243885993958 lr:5.4433555010418245e-06 tokens_per_second_per_gpu:3345.717039940193 
-Step 385 | loss:0.8795440196990967 lr:5.407920797796467e-06 tokens_per_second_per_gpu:3480.7882876326375 
-Step 386 | loss:0.8486223220825195 lr:5.372536662492775e-06 tokens_per_second_per_gpu:2192.7935593406287 
-Step 387 | loss:0.7376230955123901 lr:5.337203950405684e-06 tokens_per_second_per_gpu:3305.03296637997 
-Step 388 | loss:0.9511170387268066 lr:5.301923515567164e-06 tokens_per_second_per_gpu:2384.3194346306673 
-Step 389 | loss:0.9646419882774353 lr:5.266696210745586e-06 tokens_per_second_per_gpu:2997.5034964349775 
-Step 390 | loss:0.8492111563682556 lr:5.231522887425101e-06 tokens_per_second_per_gpu:2428.3334671595767 
-Step 391 | loss:0.8491449952125549 lr:5.196404395785076e-06 tokens_per_second_per_gpu:3138.2991290180353 
-Step 392 | loss:0.8455160856246948 lr:5.161341584679528e-06 tokens_per_second_per_gpu:2930.049873705843 
-Step 393 | loss:0.7958143949508667 lr:5.126335301616613e-06 tokens_per_second_per_gpu:2856.2636955887865 
-Step 394 | loss:0.7903842926025391 lr:5.091386392738142e-06 tokens_per_second_per_gpu:2432.8871043673084 
-Step 395 | loss:0.9168512225151062 lr:5.056495702799119e-06 tokens_per_second_per_gpu:2128.874388217258 
-Step 396 | loss:0.8486897349357605 lr:5.02166407514734e-06 tokens_per_second_per_gpu:2714.435543410219 
-Step 397 | loss:0.8215017318725586 lr:4.9868923517029894e-06 tokens_per_second_per_gpu:3343.137622210726 
-Step 398 | loss:0.8552722930908203 lr:4.9521813729383045e-06 tokens_per_second_per_gpu:2519.8796099196766 
-Step 399 | loss:0.788427472114563 lr:4.9175319778572534e-06 tokens_per_second_per_gpu:2166.8862532052426 
-Step 400 | loss:0.9515634179115295 lr:4.882945003975251e-06 tokens_per_second_per_gpu:3065.2233860728775 
-Step 401 | loss:0.810390293598175 lr:4.848421287298924e-06 tokens_per_second_per_gpu:2659.3075028331896 
-Step 402 | loss:0.8705076575279236 lr:4.813961662305905e-06 tokens_per_second_per_gpu:2782.8155438173562 
-Step 403 | loss:0.9843229055404663 lr:4.779566961924652e-06 tokens_per_second_per_gpu:2582.5921207447886 
-Step 404 | loss:0.9581823945045471 lr:4.745238017514326e-06 tokens_per_second_per_gpu:2686.0533558342913 
-Step 405 | loss:0.8852664232254028 lr:4.71097565884468e-06 tokens_per_second_per_gpu:3322.62504395818 
-Step 406 | loss:1.0313615798950195 lr:4.676780714076024e-06 tokens_per_second_per_gpu:2349.1757169557204 
-Step 407 | loss:0.8631855845451355 lr:4.6426540097391955e-06 tokens_per_second_per_gpu:2534.359572679826 
-Step 408 | loss:0.829632580280304 lr:4.6085963707155815e-06 tokens_per_second_per_gpu:2800.40334172245 
-Step 409 | loss:0.9864691495895386 lr:4.574608620217182e-06 tokens_per_second_per_gpu:2304.7700393033388 
-Step 410 | loss:0.9433596134185791 lr:4.54069157976671e-06 tokens_per_second_per_gpu:2128.7283080192083 
-Step 411 | loss:0.8118562698364258 lr:4.506846069177737e-06 tokens_per_second_per_gpu:3097.3892470665614 
-Step 412 | loss:0.8059877157211304 lr:4.473072906534878e-06 tokens_per_second_per_gpu:3564.4255635926534 
-Step 413 | loss:0.9641546010971069 lr:4.4393729081740125e-06 tokens_per_second_per_gpu:2130.3184632629923 
-Step 414 | loss:0.8071569204330444 lr:4.405746888662562e-06 tokens_per_second_per_gpu:2281.9392502990377 
-Step 415 | loss:0.8968546986579895 lr:4.372195660779782e-06 tokens_per_second_per_gpu:2232.3985049022785 
-Step 416 | loss:0.8468106389045715 lr:4.338720035497139e-06 tokens_per_second_per_gpu:3083.9242151564563 
-Step 417 | loss:0.8717325925827026 lr:4.305320821958703e-06 tokens_per_second_per_gpu:3372.5584573988763 
-Step 418 | loss:0.8747634887695312 lr:4.271998827461571e-06 tokens_per_second_per_gpu:2445.4055033972445 
-Step 419 | loss:0.9820618033409119 lr:4.238754857436381e-06 tokens_per_second_per_gpu:2342.6436446309026 
-Step 420 | loss:0.8231772780418396 lr:4.20558971542782e-06 tokens_per_second_per_gpu:2528.9826899571935 
-Step 421 | loss:0.7924979329109192 lr:4.172504203075216e-06 tokens_per_second_per_gpu:2729.9206992294567 
-Step 422 | loss:0.9110404849052429 lr:4.139499120093161e-06 tokens_per_second_per_gpu:3315.3490158265113 
-Step 423 | loss:0.75404953956604 lr:4.106575264252172e-06 tokens_per_second_per_gpu:3090.977701701262 
-Step 424 | loss:0.938966691493988 lr:4.073733431359421e-06 tokens_per_second_per_gpu:2904.2331904283437 
-Step 425 | loss:0.8179991245269775 lr:4.040974415239475e-06 tokens_per_second_per_gpu:2990.000297168357 
-Step 426 | loss:1.0704540014266968 lr:4.0082990077151445e-06 tokens_per_second_per_gpu:1991.6900351358886 
-Step 427 | loss:0.8314566016197205 lr:3.975707998588312e-06 tokens_per_second_per_gpu:3039.0039030063976 
-Step 428 | loss:0.8255379796028137 lr:3.9432021756208656e-06 tokens_per_second_per_gpu:2817.339944564254 
-Step 429 | loss:0.8793909549713135 lr:3.910782324515634e-06 tokens_per_second_per_gpu:2784.943798011668 
-Step 430 | loss:0.9148744940757751 lr:3.878449228897414e-06 tokens_per_second_per_gpu:3249.35877414692 
-Step 431 | loss:0.7125563025474548 lr:3.846203670294028e-06 tokens_per_second_per_gpu:3253.7632080442154 
-Step 432 | loss:0.9870259165763855 lr:3.814046428117417e-06 tokens_per_second_per_gpu:2434.1776382820162 
-Step 433 | loss:0.9864261150360107 lr:3.781978279644829e-06 tokens_per_second_per_gpu:2380.6335664694293 
-Step 434 | loss:0.8019827604293823 lr:3.750000000000002e-06 tokens_per_second_per_gpu:3123.600926175111 
-Step 435 | loss:0.7339569926261902 lr:3.718112362134444e-06 tokens_per_second_per_gpu:3781.257392588865 
-Step 436 | loss:0.8225141763687134 lr:3.686316136808756e-06 tokens_per_second_per_gpu:2818.7608784896115 
-Step 437 | loss:0.9751465916633606 lr:3.6546120925739825e-06 tokens_per_second_per_gpu:2280.5290359147593 
-Step 438 | loss:1.2062530517578125 lr:3.6230009957530574e-06 tokens_per_second_per_gpu:2214.766667892836 
-Step 439 | loss:0.8194683194160461 lr:3.59148361042226e-06 tokens_per_second_per_gpu:2587.946850567925 
-Step 440 | loss:0.9677116274833679 lr:3.5600606983927607e-06 tokens_per_second_per_gpu:2214.6139566548404 
-Step 441 | loss:0.8907654285430908 lr:3.528733019192198e-06 tokens_per_second_per_gpu:2491.138869734744 
-Step 442 | loss:0.8188056349754333 lr:3.4975013300463318e-06 tokens_per_second_per_gpu:2672.3598495643787 
-Step 443 | loss:0.9130949378013611 lr:3.4663663858607234e-06 tokens_per_second_per_gpu:2951.3410891423564 
-Step 444 | loss:0.9273456335067749 lr:3.4353289392025103e-06 tokens_per_second_per_gpu:2988.781719929555 
-Step 445 | loss:0.7225141525268555 lr:3.404389740282191e-06 tokens_per_second_per_gpu:3436.804380837058 
-Step 446 | loss:0.8584692478179932 lr:3.3735495369355077e-06 tokens_per_second_per_gpu:3279.7335439817807 
-Step 447 | loss:0.8927696347236633 lr:3.342809074605375e-06 tokens_per_second_per_gpu:2235.93741901587 
-Step 448 | loss:0.9940071105957031 lr:3.3121690963238414e-06 tokens_per_second_per_gpu:2386.527757758584 
-Step 449 | loss:0.7678663730621338 lr:3.2816303426941454e-06 tokens_per_second_per_gpu:3482.820850954397 
-Step 450 | loss:0.9011093378067017 lr:3.2511935518728157e-06 tokens_per_second_per_gpu:1981.403319469987 
-Step 451 | loss:0.9154335260391235 lr:3.2208594595518152e-06 tokens_per_second_per_gpu:2839.676474707793 
-Step 452 | loss:0.9392225742340088 lr:3.1906287989407737e-06 tokens_per_second_per_gpu:2792.9917171852126 
-Step 453 | loss:0.941868245601654 lr:3.1605023007492552e-06 tokens_per_second_per_gpu:2564.1183459402805 
-Step 454 | loss:0.8091152310371399 lr:3.1304806931690975e-06 tokens_per_second_per_gpu:3746.246185380797 
-Step 455 | loss:0.9475477933883667 lr:3.100564701856814e-06 tokens_per_second_per_gpu:2413.935688483254 
-Step 456 | loss:0.9426826238632202 lr:3.070755049916057e-06 tokens_per_second_per_gpu:2303.012048381734 
-Step 457 | loss:1.0177905559539795 lr:3.0410524578801263e-06 tokens_per_second_per_gpu:2555.0729947502355 
-Step 458 | loss:0.9545515775680542 lr:3.01145764369457e-06 tokens_per_second_per_gpu:2975.888591278216 
-Step 459 | loss:0.8237252831459045 lr:2.9819713226998186e-06 tokens_per_second_per_gpu:2493.940474143623 
-Step 460 | loss:0.963207483291626 lr:2.9525942076138937e-06 tokens_per_second_per_gpu:1983.0426983275186 
-Step 461 | loss:0.9316860437393188 lr:2.9233270085151965e-06 tokens_per_second_per_gpu:2681.451947505957 
-Step 462 | loss:0.880352258682251 lr:2.8941704328253223e-06 tokens_per_second_per_gpu:2415.10368731483 
-Step 463 | loss:1.0479156970977783 lr:2.8651251852919812e-06 tokens_per_second_per_gpu:2803.8099792533726 
-Step 464 | loss:0.902858555316925 lr:2.8361919679719494e-06 tokens_per_second_per_gpu:2516.2445296751976 
-Step 465 | loss:1.0508098602294922 lr:2.8073714802141027e-06 tokens_per_second_per_gpu:2480.442905915016 
-Step 466 | loss:0.8653034567832947 lr:2.7786644186425245e-06 tokens_per_second_per_gpu:1932.6317201110714 
-Step 467 | loss:0.9926808476448059 lr:2.7500714771396464e-06 tokens_per_second_per_gpu:1877.5562851011355 
-Step 468 | loss:0.8932638168334961 lr:2.721593346829501e-06 tokens_per_second_per_gpu:3161.1173865778374 
-Step 469 | loss:0.8979821801185608 lr:2.693230716060984e-06 tokens_per_second_per_gpu:2036.8558420047852 
-Step 470 | loss:0.7667185664176941 lr:2.6649842703912558e-06 tokens_per_second_per_gpu:2900.220984639236 
-Step 471 | loss:0.7836153507232666 lr:2.636854692569135e-06 tokens_per_second_per_gpu:2817.8863131727257 
-Step 472 | loss:0.9054886102676392 lr:2.6088426625186216e-06 tokens_per_second_per_gpu:2930.6798374473015 
-Step 473 | loss:0.860165536403656 lr:2.58094885732245e-06 tokens_per_second_per_gpu:2410.5742824676117 
-Step 474 | loss:0.9614354968070984 lr:2.553173951205715e-06 tokens_per_second_per_gpu:2776.494678342581 
-Step 475 | loss:0.8513482809066772 lr:2.5255186155195993e-06 tokens_per_second_per_gpu:3631.276628324878 
-Step 476 | loss:0.8641606569290161 lr:2.497983518725119e-06 tokens_per_second_per_gpu:2638.6224491605603 
-Step 477 | loss:0.8530397415161133 lr:2.470569326376991e-06 tokens_per_second_per_gpu:2284.1034573050524 
-Step 478 | loss:0.8508715033531189 lr:2.443276701107525e-06 tokens_per_second_per_gpu:2732.368505701632 
-Step 479 | loss:0.8788378834724426 lr:2.4161063026106153e-06 tokens_per_second_per_gpu:2150.624908574507 
-Step 480 | loss:0.962777316570282 lr:2.389058787625805e-06 tokens_per_second_per_gpu:2568.8995848617747 
-Step 481 | loss:0.7333717942237854 lr:2.3621348099223907e-06 tokens_per_second_per_gpu:3710.6072079951878 
-Step 482 | loss:0.9811513423919678 lr:2.335335020283642e-06 tokens_per_second_per_gpu:2925.5796188715262 
-Step 483 | loss:0.8781923651695251 lr:2.308660066491055e-06 tokens_per_second_per_gpu:2126.7808277712015 
-Step 484 | loss:0.9604898691177368 lr:2.282110593308702e-06 tokens_per_second_per_gpu:1926.7847478492406 
-Step 485 | loss:1.1256814002990723 lr:2.2556872424676447e-06 tokens_per_second_per_gpu:2558.8925285932874 
-Step 486 | loss:0.9171305894851685 lr:2.22939065265043e-06 tokens_per_second_per_gpu:1903.1195805184752 
-Step 487 | loss:0.8313721418380737 lr:2.203221459475638e-06 tokens_per_second_per_gpu:2979.98771275172 
-Step 488 | loss:0.8055751323699951 lr:2.1771802954825338e-06 tokens_per_second_per_gpu:3015.6667404412465 
-Step 489 | loss:1.0302447080612183 lr:2.151267790115766e-06 tokens_per_second_per_gpu:2216.9913058020384 
-Step 490 | loss:1.1314735412597656 lr:2.1254845697101576e-06 tokens_per_second_per_gpu:2104.5816237570134 
-Step 491 | loss:0.9303261637687683 lr:2.099831257475571e-06 tokens_per_second_per_gpu:2165.4810755645126 
-Step 492 | loss:0.7989121675491333 lr:2.074308473481833e-06 tokens_per_second_per_gpu:2985.2464129518576 
-Step 493 | loss:1.00505530834198 lr:2.0489168346437635e-06 tokens_per_second_per_gpu:2471.4941659452065 
-Step 494 | loss:1.0171761512756348 lr:2.023656954706244e-06 tokens_per_second_per_gpu:2918.608675039668 
-Step 495 | loss:0.937871515750885 lr:1.998529444229398e-06 tokens_per_second_per_gpu:2691.00456421299 
-Step 496 | loss:0.8979736566543579 lr:1.97353491057383e-06 tokens_per_second_per_gpu:2747.281150085522 
-Step 497 | loss:0.8915017247200012 lr:1.948673957885937e-06 tokens_per_second_per_gpu:2392.395426595406 
-Step 498 | loss:0.9766404628753662 lr:1.923947187083323e-06 tokens_per_second_per_gpu:2406.2149686418447 
-Step 499 | loss:0.903343915939331 lr:1.8993551958402444e-06 tokens_per_second_per_gpu:3398.4092182111635 
-Step 500 | loss:0.8752915263175964 lr:1.8748985785732007e-06 tokens_per_second_per_gpu:2725.839322744673 
-Step 501 | loss:0.9260117411613464 lr:1.8505779264265335e-06 tokens_per_second_per_gpu:2492.531712956765 
-Step 502 | loss:0.8971195220947266 lr:1.8263938272581646e-06 tokens_per_second_per_gpu:2901.933455496639 
-Step 503 | loss:0.8458098769187927 lr:1.8023468656253648e-06 tokens_per_second_per_gpu:2507.8217453974867 
-Step 504 | loss:1.0840578079223633 lr:1.7784376227706355e-06 tokens_per_second_per_gpu:2613.3231747806803 
-Step 505 | loss:0.9679098725318909 lr:1.7546666766076658e-06 tokens_per_second_per_gpu:2675.238641139705 
-Step 506 | loss:0.879393994808197 lr:1.7310346017073452e-06 tokens_per_second_per_gpu:2837.0564466992423 
-Step 507 | loss:0.7026180624961853 lr:1.7075419692838958e-06 tokens_per_second_per_gpu:2933.1935810857617 
-Step 508 | loss:0.848726212978363 lr:1.6841893471810504e-06 tokens_per_second_per_gpu:2718.772584505565 
-Step 509 | loss:0.9221838712692261 lr:1.6609772998583307e-06 tokens_per_second_per_gpu:1915.4857629787555 
-Step 510 | loss:0.8497966527938843 lr:1.6379063883774126e-06 tokens_per_second_per_gpu:2039.9625666861666 
-Step 511 | loss:0.8197170495986938 lr:1.6149771703885507e-06 tokens_per_second_per_gpu:2976.637056029542 
-Step 512 | loss:0.8008678555488586 lr:1.5921902001171148e-06 tokens_per_second_per_gpu:3667.89278300197 
-Step 513 | loss:0.9182004332542419 lr:1.5695460283501717e-06 tokens_per_second_per_gpu:2733.1730924705953 
-Step 514 | loss:0.9430605173110962 lr:1.5470452024231982e-06 tokens_per_second_per_gpu:2897.3090684371555 
-Step 515 | loss:0.7945107817649841 lr:1.5246882662068304e-06 tokens_per_second_per_gpu:3425.0149827938044 
-Step 516 | loss:0.9675235748291016 lr:1.5024757600937314e-06 tokens_per_second_per_gpu:2672.878682305898 
-Step 517 | loss:0.8425970673561096 lr:1.4804082209855243e-06 tokens_per_second_per_gpu:2473.8463345979508 
-Step 518 | loss:0.8976823687553406 lr:1.4584861822798037e-06 tokens_per_second_per_gpu:2927.3740187446383 
-Step 519 | loss:0.9743514657020569 lr:1.4367101738572688e-06 tokens_per_second_per_gpu:2356.6379119384883 
-Step 520 | loss:0.8489241600036621 lr:1.4150807220688866e-06 tokens_per_second_per_gpu:2448.2238090330206 
-Step 521 | loss:0.7413209676742554 lr:1.3935983497231943e-06 tokens_per_second_per_gpu:3176.6007732610974 
-Step 522 | loss:0.7783915400505066 lr:1.3722635760736431e-06 tokens_per_second_per_gpu:3504.9946590495997 
-Step 523 | loss:0.8300887942314148 lr:1.351076916806057e-06 tokens_per_second_per_gpu:2424.9390276385934 
-Step 524 | loss:0.9937973618507385 lr:1.3300388840261708e-06 tokens_per_second_per_gpu:2906.0669261298367 
-Step 525 | loss:0.9671134948730469 lr:1.3091499862472395e-06 tokens_per_second_per_gpu:2832.7584552787275 
-Step 526 | loss:0.9223825931549072 lr:1.2884107283777644e-06 tokens_per_second_per_gpu:2943.4211173912004 
-Step 527 | loss:0.911368727684021 lr:1.2678216117092729e-06 tokens_per_second_per_gpu:3176.3560024048184 
-Step 528 | loss:0.9336380958557129 lr:1.2473831339042085e-06 tokens_per_second_per_gpu:2298.0152377434993 
-Step 529 | loss:0.8240398168563843 lr:1.227095788983903e-06 tokens_per_second_per_gpu:2798.8999239376394 
-Step 530 | loss:0.7940638065338135 lr:1.206960067316636e-06 tokens_per_second_per_gpu:3165.2590382202575 
-Step 531 | loss:0.8540267944335938 lr:1.186976455605775e-06 tokens_per_second_per_gpu:2187.2335292148864 
-Step 532 | loss:0.8410139083862305 lr:1.1671454368780222e-06 tokens_per_second_per_gpu:2474.942310509921 
-Step 533 | loss:1.1205919981002808 lr:1.1474674904717284e-06 tokens_per_second_per_gpu:2181.678986703241 
-Step 534 | loss:0.8875789642333984 lr:1.1279430920253122e-06 tokens_per_second_per_gpu:2593.2963355514185 
-Step 535 | loss:0.9577472805976868 lr:1.1085727134657672e-06 tokens_per_second_per_gpu:2744.679461314659 
-Step 536 | loss:0.7591468691825867 lr:1.0893568229972445e-06 tokens_per_second_per_gpu:2984.9695645483403 
-Step 537 | loss:0.8278703093528748 lr:1.0702958850897482e-06 tokens_per_second_per_gpu:3224.3677192462064 
-Step 538 | loss:0.9749331474304199 lr:1.0513903604678959e-06 tokens_per_second_per_gpu:2179.7933704651227 
-Step 539 | loss:0.8200403451919556 lr:1.0326407060997912e-06 tokens_per_second_per_gpu:2585.873381589468 
-Step 540 | loss:0.8951087594032288 lr:1.0140473751859792e-06 tokens_per_second_per_gpu:2359.7089376063186 
-Step 541 | loss:0.8001818060874939 lr:9.956108171484813e-07 tokens_per_second_per_gpu:3679.2858057381336 
-Step 542 | loss:0.9417346119880676 lr:9.773314776199499e-07 tokens_per_second_per_gpu:3021.156241879852 
-Step 543 | loss:0.7579294443130493 lr:9.59209798432874e-07 tokens_per_second_per_gpu:2599.789473685248 
-Step 544 | loss:0.9392054080963135 lr:9.412462176089243e-07 tokens_per_second_per_gpu:2654.3657026369724 
-Step 545 | loss:1.048731803894043 lr:9.234411693483439e-07 tokens_per_second_per_gpu:2791.9461967353373 
-Step 546 | loss:0.8341087102890015 lr:9.057950840194687e-07 tokens_per_second_per_gpu:2785.6629892119927 
-Step 547 | loss:0.8932492136955261 lr:8.883083881483225e-07 tokens_per_second_per_gpu:2820.3743680286916 
-Step 548 | loss:0.922091543674469 lr:8.709815044082916e-07 tokens_per_second_per_gpu:2609.799256644826 
-Step 549 | loss:0.8205937147140503 lr:8.53814851609934e-07 tokens_per_second_per_gpu:2846.3717400831083 
-Step 550 | loss:0.7681778073310852 lr:8.368088446908337e-07 tokens_per_second_per_gpu:3047.900830971702 
-Step 551 | loss:1.028059720993042 lr:8.199638947055871e-07 tokens_per_second_per_gpu:2346.733057129004 
-Step 552 | loss:0.9609188437461853 lr:8.032804088158568e-07 tokens_per_second_per_gpu:2578.2629488621687 
-Step 553 | loss:0.8993929028511047 lr:7.867587902805334e-07 tokens_per_second_per_gpu:2471.7660146749104 
-Step 554 | loss:0.9311540126800537 lr:7.703994384459929e-07 tokens_per_second_per_gpu:2232.6221368289566 
-Step 555 | loss:1.0247182846069336 lr:7.542027487364342e-07 tokens_per_second_per_gpu:2030.4209899077885 
-Step 556 | loss:0.8552548885345459 lr:7.381691126443343e-07 tokens_per_second_per_gpu:2798.180824884108 
-Step 557 | loss:0.8369226455688477 lr:7.222989177209718e-07 tokens_per_second_per_gpu:1813.2487757380513 
-Step 558 | loss:0.9284541010856628 lr:7.065925475670679e-07 tokens_per_second_per_gpu:2572.308551755442 
-Step 559 | loss:1.1530259847640991 lr:6.910503818235119e-07 tokens_per_second_per_gpu:1981.7748187521522 
-Step 560 | loss:0.8481274843215942 lr:6.75672796162187e-07 tokens_per_second_per_gpu:2449.174249102278 
-Step 561 | loss:1.0015965700149536 lr:6.604601622768886e-07 tokens_per_second_per_gpu:2805.9037697407794 
-Step 562 | loss:0.8001527786254883 lr:6.454128478743318e-07 tokens_per_second_per_gpu:3317.809001429568 
-Step 563 | loss:0.78518146276474 lr:6.305312166652816e-07 tokens_per_second_per_gpu:3155.550225666974 
-Step 564 | loss:0.875720739364624 lr:6.158156283557429e-07 tokens_per_second_per_gpu:2783.8704022847533 
-Step 565 | loss:0.9692037105560303 lr:6.01266438638281e-07 tokens_per_second_per_gpu:2805.3746426385587 
-Step 566 | loss:1.054160237312317 lr:5.868839991834129e-07 tokens_per_second_per_gpu:2411.8992377023055 
-Step 567 | loss:0.9159079194068909 lr:5.726686576311104e-07 tokens_per_second_per_gpu:2706.8539397659056 
-Step 568 | loss:0.8968900442123413 lr:5.586207575824046e-07 tokens_per_second_per_gpu:3760.640807850762 
-Step 569 | loss:0.8418700098991394 lr:5.447406385910661e-07 tokens_per_second_per_gpu:2949.5486193635475 
-Step 570 | loss:1.0156406164169312 lr:5.310286361554125e-07 tokens_per_second_per_gpu:2600.276591475816 
-Step 571 | loss:0.825329601764679 lr:5.174850817101903e-07 tokens_per_second_per_gpu:2637.289992319456 
-Step 572 | loss:0.9666601419448853 lr:5.041103026185642e-07 tokens_per_second_per_gpu:2454.7473876850663 
-Step 573 | loss:0.6319097280502319 lr:4.909046221642066e-07 tokens_per_second_per_gpu:4199.945046520476 
-Step 574 | loss:0.8879971504211426 lr:4.77868359543486e-07 tokens_per_second_per_gpu:2809.1533118309153 
-Step 575 | loss:0.8433800935745239 lr:4.6500182985774406e-07 tokens_per_second_per_gpu:3222.208727332213 
-Step 576 | loss:0.7500202655792236 lr:4.5230534410568764e-07 tokens_per_second_per_gpu:3835.5875929096846 
-Step 577 | loss:0.8776817321777344 lr:4.3977920917586455e-07 tokens_per_second_per_gpu:3588.3562798100056 
-Step 578 | loss:0.7789593935012817 lr:4.2742372783924984e-07 tokens_per_second_per_gpu:2702.664818633485 
-Step 579 | loss:1.0058625936508179 lr:4.1523919874192797e-07 tokens_per_second_per_gpu:3198.8889390277964 
-Step 580 | loss:1.1509525775909424 lr:4.0322591639786885e-07 tokens_per_second_per_gpu:1779.936655769417 
-Step 581 | loss:0.8506143093109131 lr:3.9138417118182e-07 tokens_per_second_per_gpu:2704.5816345540325 
-Step 582 | loss:1.063564419746399 lr:3.797142493222733e-07 tokens_per_second_per_gpu:1961.6927563427632 
-Step 583 | loss:1.075540542602539 lr:3.6821643289455586e-07 tokens_per_second_per_gpu:2343.735008970859 
-Step 584 | loss:0.9307299852371216 lr:3.5689099981401517e-07 tokens_per_second_per_gpu:2528.258117529406 
-Step 585 | loss:0.909044086933136 lr:3.457382238292892e-07 tokens_per_second_per_gpu:2716.7367247826746 
-Step 586 | loss:0.9028974771499634 lr:3.3475837451570404e-07 tokens_per_second_per_gpu:2008.4029901200033 
-Step 587 | loss:0.8828522562980652 lr:3.239517172687428e-07 tokens_per_second_per_gpu:2328.6354197459186 
-Step 588 | loss:0.9550468921661377 lr:3.1331851329764654e-07 tokens_per_second_per_gpu:2858.7367313426307 
-Step 589 | loss:0.964181661605835 lr:3.028590196190864e-07 tokens_per_second_per_gpu:2039.593965890504 
-Step 590 | loss:0.8647264838218689 lr:2.9257348905096156e-07 tokens_per_second_per_gpu:2408.2528550609186 
-Step 591 | loss:0.7582616209983826 lr:2.8246217020628544e-07 tokens_per_second_per_gpu:3584.2282370995854 
-Step 592 | loss:0.9030933380126953 lr:2.725253074871692e-07 tokens_per_second_per_gpu:2740.8135271861065 
-Step 593 | loss:0.820608913898468 lr:2.6276314107892607e-07 tokens_per_second_per_gpu:2632.8270224856783 
-Step 594 | loss:0.9622331261634827 lr:2.5317590694425716e-07 tokens_per_second_per_gpu:2733.3939723230974 
-Step 595 | loss:0.8576785326004028 lr:2.4376383681755323e-07 tokens_per_second_per_gpu:2548.113990591433 
-Step 596 | loss:0.8732478022575378 lr:2.345271581992886e-07 tokens_per_second_per_gpu:2347.85222649422 
-Step 597 | loss:0.9593178033828735 lr:2.2546609435052546e-07 tokens_per_second_per_gpu:2727.4039541246357 
-Step 598 | loss:0.9085021018981934 lr:2.1658086428751888e-07 tokens_per_second_per_gpu:2564.083915396395 
-Step 599 | loss:0.9484013915061951 lr:2.0787168277641627e-07 tokens_per_second_per_gpu:2287.069566160566 
-Step 600 | loss:0.7718688249588013 lr:1.993387603280755e-07 tokens_per_second_per_gpu:2726.9851612511115 
-Step 601 | loss:0.9098013043403625 lr:1.9098230319296834e-07 tokens_per_second_per_gpu:1928.3522236853428 
-Step 602 | loss:0.8552864789962769 lr:1.828025133561992e-07 tokens_per_second_per_gpu:2926.7052991542437 
-Step 603 | loss:0.9261455535888672 lr:1.747995885326234e-07 tokens_per_second_per_gpu:3406.735424527618 
-Step 604 | loss:0.9243958592414856 lr:1.6697372216206514e-07 tokens_per_second_per_gpu:2535.3725387216314 
-Step 605 | loss:0.9653189182281494 lr:1.5932510340464694e-07 tokens_per_second_per_gpu:2746.3850646471515 
-Step 606 | loss:0.834320604801178 lr:1.51853917136211e-07 tokens_per_second_per_gpu:3429.3423919016786 
-Step 607 | loss:0.8928940892219543 lr:1.4456034394385587e-07 tokens_per_second_per_gpu:2297.879477800558 
-Step 608 | loss:0.7676761746406555 lr:1.3744456012156764e-07 tokens_per_second_per_gpu:3314.490893171432 
-Step 609 | loss:0.8587438464164734 lr:1.305067376659616e-07 tokens_per_second_per_gpu:2284.4486288898347 
-Step 610 | loss:1.0006862878799438 lr:1.2374704427212392e-07 tokens_per_second_per_gpu:2558.2291684611323 
-Step 611 | loss:0.9732027649879456 lr:1.1716564332955815e-07 tokens_per_second_per_gpu:3952.588941664968 
-Step 612 | loss:1.2337478399276733 lr:1.1076269391823435e-07 tokens_per_second_per_gpu:1877.1004881061126 
-Step 613 | loss:0.8134404420852661 lr:1.0453835080474706e-07 tokens_per_second_per_gpu:3074.0139322158097 
-Step 614 | loss:1.006320834159851 lr:9.849276443857164e-08 tokens_per_second_per_gpu:2227.792011738089 
-Step 615 | loss:0.8254751563072205 lr:9.262608094842972e-08 tokens_per_second_per_gpu:2500.2796254751793 
-Step 616 | loss:0.8692588806152344 lr:8.693844213875702e-08 tokens_per_second_per_gpu:2593.9356798742424 
-Step 617 | loss:0.9400972723960876 lr:8.142998548627184e-08 tokens_per_second_per_gpu:2395.1619591858566 
-Step 618 | loss:0.9176926016807556 lr:7.610084413665869e-08 tokens_per_second_per_gpu:3514.4466872465096 
-Step 619 | loss:0.812978208065033 lr:7.095114690134574e-08 tokens_per_second_per_gpu:2740.6624145248784 
-Step 620 | loss:0.8113747835159302 lr:6.598101825439074e-08 tokens_per_second_per_gpu:2798.1631566407473 
-Step 621 | loss:0.9036160707473755 lr:6.1190578329475e-08 tokens_per_second_per_gpu:2909.508395066856 
-Step 622 | loss:0.9566043615341187 lr:5.6579942916997474e-08 tokens_per_second_per_gpu:2437.719725089656 
-Step 623 | loss:0.8796412348747253 lr:5.214922346127859e-08 tokens_per_second_per_gpu:2324.519840460166 
-Step 624 | loss:0.861414909362793 lr:4.7898527057862464e-08 tokens_per_second_per_gpu:2770.5794749156607 
-Step 625 | loss:0.8064419031143188 lr:4.382795645093307e-08 tokens_per_second_per_gpu:2469.646669989894 
-Step 626 | loss:0.9710454940795898 lr:3.9937610030827134e-08 tokens_per_second_per_gpu:2404.1522318995067 
-Step 627 | loss:0.881490170955658 lr:3.6227581831655144e-08 tokens_per_second_per_gpu:2572.367154951392 
-Step 628 | loss:0.82835853099823 lr:3.269796152903404e-08 tokens_per_second_per_gpu:3620.7784223349213 
-Step 629 | loss:0.8445915579795837 lr:2.934883443791392e-08 tokens_per_second_per_gpu:2854.4862142352713 
-Step 630 | loss:1.0001657009124756 lr:2.6180281510518044e-08 tokens_per_second_per_gpu:2978.064983348182 
-Step 631 | loss:0.9440408945083618 lr:2.3192379334386894e-08 tokens_per_second_per_gpu:2696.540350146697 
-Step 632 | loss:0.7228038907051086 lr:2.0385200130525472e-08 tokens_per_second_per_gpu:3655.488183918562 
-Step 633 | loss:0.9987472891807556 lr:1.7758811751658887e-08 tokens_per_second_per_gpu:2617.6189994415217 
-Step 634 | loss:0.8774012923240662 lr:1.531327768059032e-08 tokens_per_second_per_gpu:2923.040995277404 
-Step 635 | loss:0.949100911617279 lr:1.3048657028669752e-08 tokens_per_second_per_gpu:2582.158040430071 
-Step 636 | loss:0.940438985824585 lr:1.096500453436261e-08 tokens_per_second_per_gpu:2001.7866179040327 
-Step 637 | loss:0.9458042979240417 lr:9.062370561927491e-09 tokens_per_second_per_gpu:2777.8153040677407 
-Step 638 | loss:0.8899053931236267 lr:7.340801100197969e-09 tokens_per_second_per_gpu:2513.1679147270943 
-Step 639 | loss:0.8548555374145508 lr:5.800337761473484e-09 tokens_per_second_per_gpu:2289.6067396171884 
-Step 640 | loss:0.9246522784233093 lr:4.441017780510148e-09 tokens_per_second_per_gpu:2645.0215660487215 
-Step 641 | loss:1.0237187147140503 lr:3.262874013622297e-09 tokens_per_second_per_gpu:2515.905628267492 
-Step 642 | loss:0.8752281665802002 lr:2.265934937891456e-09 tokens_per_second_per_gpu:2979.220301682006 
-Step 643 | loss:0.962019681930542 lr:1.450224650471066e-09 tokens_per_second_per_gpu:2538.88920163452 
-Step 644 | loss:0.8064459562301636 lr:8.157628680094397e-10 tokens_per_second_per_gpu:2978.4258421814507 
-Step 645 | loss:0.8946848511695862 lr:3.6256492617264826e-10 tokens_per_second_per_gpu:2379.746472667621 
-Step 646 | loss:0.7861452102661133 lr:9.064177927231576e-11 tokens_per_second_per_gpu:2941.276869680707 
-Step 647 | loss:0.9995426535606384 lr:0.0 tokens_per_second_per_gpu:2276.393936517497 
+Step 1 | loss:1.0854802131652832 lr:1.875e-06 tokens_per_second_per_gpu:1168.8574294071448 
+Step 2 | loss:1.1429901123046875 lr:3.75e-06 tokens_per_second_per_gpu:3843.7632889854767 
+Step 3 | loss:1.2053295373916626 lr:5.625e-06 tokens_per_second_per_gpu:3842.159382826307 
+Step 4 | loss:1.079384446144104 lr:7.5e-06 tokens_per_second_per_gpu:3527.3851566722074 
+Step 5 | loss:1.2358113527297974 lr:9.375000000000001e-06 tokens_per_second_per_gpu:3297.665429778813 
+Step 6 | loss:1.0729756355285645 lr:1.125e-05 tokens_per_second_per_gpu:3513.7372696945217 
+Step 7 | loss:0.9907684326171875 lr:1.3125e-05 tokens_per_second_per_gpu:3930.22637976525 
+Step 8 | loss:1.143568515777588 lr:1.5e-05 tokens_per_second_per_gpu:2876.219047245818 
+Step 9 | loss:1.2526814937591553 lr:1.4999908501094575e-05 tokens_per_second_per_gpu:3772.480582082779 
+Step 10 | loss:1.1926498413085938 lr:1.4999634006610844e-05 tokens_per_second_per_gpu:3787.9663127264835 
+Step 11 | loss:1.1863847970962524 lr:1.4999176523246392e-05 tokens_per_second_per_gpu:2879.4265754703606 
+Step 12 | loss:1.0382694005966187 lr:1.4998536062163683e-05 tokens_per_second_per_gpu:3367.342512712612 
+Step 13 | loss:1.0486845970153809 lr:1.4997712638989775e-05 tokens_per_second_per_gpu:3973.1834061223685 
+Step 14 | loss:1.0994930267333984 lr:1.499670627381596e-05 tokens_per_second_per_gpu:3882.9148296055278 
+Step 15 | loss:1.0168027877807617 lr:1.4995516991197246e-05 tokens_per_second_per_gpu:3710.4998349202683 
+Step 16 | loss:1.0419063568115234 lr:1.4994144820151789e-05 tokens_per_second_per_gpu:2869.5305816393893 
+Step 17 | loss:1.065361738204956 lr:1.4992589794160159e-05 tokens_per_second_per_gpu:3181.270854761355 
+Step 18 | loss:1.1389403343200684 lr:1.4990851951164537e-05 tokens_per_second_per_gpu:2980.579149781292 
+Step 19 | loss:1.1638612747192383 lr:1.4988931333567785e-05 tokens_per_second_per_gpu:3509.8801697983517 
+Step 20 | loss:1.0619863271713257 lr:1.4986827988232412e-05 tokens_per_second_per_gpu:3703.804663132431 
+Step 21 | loss:1.1482210159301758 lr:1.4984541966479427e-05 tokens_per_second_per_gpu:3560.136617406801 
+Step 22 | loss:1.0800286531448364 lr:1.4982073324087097e-05 tokens_per_second_per_gpu:3648.6607162752957 
+Step 23 | loss:1.0815565586090088 lr:1.4979422121289576e-05 tokens_per_second_per_gpu:3211.609323443709 
+Step 24 | loss:1.0016083717346191 lr:1.4976588422775437e-05 tokens_per_second_per_gpu:3403.8935665137487 
+Step 25 | loss:1.054980993270874 lr:1.4973572297686097e-05 tokens_per_second_per_gpu:3733.187288468551 
+Step 26 | loss:1.0594943761825562 lr:1.4970373819614128e-05 tokens_per_second_per_gpu:3380.227950545309 
+Step 27 | loss:1.1039057970046997 lr:1.4966993066601459e-05 tokens_per_second_per_gpu:3416.5838997236365 
+Step 28 | loss:1.0442756414413452 lr:1.4963430121137482e-05 tokens_per_second_per_gpu:3232.857948752556 
+Step 29 | loss:1.3250808715820312 lr:1.495968507015702e-05 tokens_per_second_per_gpu:2739.8843861816085 
+Step 30 | loss:1.0642691850662231 lr:1.495575800503823e-05 tokens_per_second_per_gpu:3270.2234151530183 
+Step 31 | loss:1.0556732416152954 lr:1.4951649021600348e-05 tokens_per_second_per_gpu:3280.5378111148802 
+Step 32 | loss:1.2651450634002686 lr:1.4947358220101378e-05 tokens_per_second_per_gpu:3675.5567883584245 
+Step 33 | loss:0.9953560829162598 lr:1.4942885705235616e-05 tokens_per_second_per_gpu:3957.394946057751 
+Step 34 | loss:0.9306876063346863 lr:1.493823158613113e-05 tokens_per_second_per_gpu:2993.4562512796037 
+Step 35 | loss:0.9759434461593628 lr:1.4933395976347056e-05 tokens_per_second_per_gpu:3688.3981271180473 
+Step 36 | loss:1.042667031288147 lr:1.4928378993870868e-05 tokens_per_second_per_gpu:2739.611697714639 
+Step 37 | loss:1.1176235675811768 lr:1.4923180761115471e-05 tokens_per_second_per_gpu:3389.9575685437558 
+Step 38 | loss:1.0304946899414062 lr:1.491780140491623e-05 tokens_per_second_per_gpu:2751.426529078761 
+Step 39 | loss:1.1554368734359741 lr:1.4912241056527865e-05 tokens_per_second_per_gpu:3678.512086747305 
+Step 40 | loss:1.1235543489456177 lr:1.4906499851621251e-05 tokens_per_second_per_gpu:3354.288583814501 
+Step 41 | loss:1.1204215288162231 lr:1.4900577930280117e-05 tokens_per_second_per_gpu:2646.7652669210534 
+Step 42 | loss:1.159371256828308 lr:1.489447543699761e-05 tokens_per_second_per_gpu:3113.893638504558 
+Step 43 | loss:0.9518178701400757 lr:1.488819252067279e-05 tokens_per_second_per_gpu:3624.505087259554 
+Step 44 | loss:1.000114917755127 lr:1.4881729334606977e-05 tokens_per_second_per_gpu:3995.207620964973 
+Step 45 | loss:1.0402061939239502 lr:1.4875086036500025e-05 tokens_per_second_per_gpu:3694.213698196683 
+Step 46 | loss:1.1420111656188965 lr:1.4868262788446472e-05 tokens_per_second_per_gpu:3655.494511670644 
+Step 47 | loss:1.0602643489837646 lr:1.4861259756931577e-05 tokens_per_second_per_gpu:3402.054859100221 
+Step 48 | loss:0.9862946271896362 lr:1.4854077112827263e-05 tokens_per_second_per_gpu:3261.120378766765 
+Step 49 | loss:1.0737448930740356 lr:1.4846715031387952e-05 tokens_per_second_per_gpu:3560.4513581368537 
+Step 50 | loss:1.0223171710968018 lr:1.483917369224628e-05 tokens_per_second_per_gpu:4119.596935775298 
+Step 51 | loss:0.9688004851341248 lr:1.4831453279408724e-05 tokens_per_second_per_gpu:3727.7426885377727 
+Step 52 | loss:1.2084407806396484 lr:1.4823553981251103e-05 tokens_per_second_per_gpu:3475.565097423413 
+Step 53 | loss:1.0782098770141602 lr:1.4815475990513983e-05 tokens_per_second_per_gpu:3116.162844911365 
+Step 54 | loss:1.076345682144165 lr:1.4807219504297984e-05 tokens_per_second_per_gpu:3636.4074969842172 
+Step 55 | loss:0.9842542409896851 lr:1.4798784724058958e-05 tokens_per_second_per_gpu:3890.3173372257284 
+Step 56 | loss:1.0145485401153564 lr:1.4790171855603081e-05 tokens_per_second_per_gpu:3849.2883562880224 
+Step 57 | loss:1.0625418424606323 lr:1.4781381109081831e-05 tokens_per_second_per_gpu:3571.1217369942487 
+Step 58 | loss:1.0252196788787842 lr:1.4772412698986854e-05 tokens_per_second_per_gpu:3456.6789270442778 
+Step 59 | loss:0.9392111897468567 lr:1.4763266844144741e-05 tokens_per_second_per_gpu:3657.3789166667807 
+Step 60 | loss:1.0814956426620483 lr:1.4753943767711678e-05 tokens_per_second_per_gpu:3315.163159791015 
+Step 61 | loss:1.0520405769348145 lr:1.4744443697168013e-05 tokens_per_second_per_gpu:3701.303137175016 
+Step 62 | loss:1.0226781368255615 lr:1.473476686431269e-05 tokens_per_second_per_gpu:3537.4955448330916 
+Step 63 | loss:1.0486294031143188 lr:1.4724913505257609e-05 tokens_per_second_per_gpu:3095.7445591899086 
+Step 64 | loss:1.0411049127578735 lr:1.4714883860421854e-05 tokens_per_second_per_gpu:3149.287379730578 
+Step 65 | loss:1.0797317028045654 lr:1.4704678174525831e-05 tokens_per_second_per_gpu:3445.673531691869 
+Step 66 | loss:0.9485515356063843 lr:1.4694296696585298e-05 tokens_per_second_per_gpu:3864.0149006705747 
+Step 67 | loss:0.9495678544044495 lr:1.4683739679905284e-05 tokens_per_second_per_gpu:3101.8256009234356 
+Step 68 | loss:1.0279844999313354 lr:1.4673007382073919e-05 tokens_per_second_per_gpu:3463.4301964845763 
+Step 69 | loss:1.0704870223999023 lr:1.4662100064956132e-05 tokens_per_second_per_gpu:3121.3618232019967 
+Step 70 | loss:1.0774767398834229 lr:1.4651017994687282e-05 tokens_per_second_per_gpu:2970.1779475300104 
+Step 71 | loss:1.040583848953247 lr:1.4639761441666646e-05 tokens_per_second_per_gpu:3368.9417024963573 
+Step 72 | loss:1.1608614921569824 lr:1.4628330680550833e-05 tokens_per_second_per_gpu:3591.1238850356212 
+Step 73 | loss:1.1760778427124023 lr:1.4616725990247078e-05 tokens_per_second_per_gpu:3191.835895149416 
+Step 74 | loss:1.0865557193756104 lr:1.4604947653906435e-05 tokens_per_second_per_gpu:3856.472486717953 
+Step 75 | loss:0.997776985168457 lr:1.4592995958916877e-05 tokens_per_second_per_gpu:4288.807801484956 
+Step 76 | loss:1.0847547054290771 lr:1.4580871196896266e-05 tokens_per_second_per_gpu:3642.7905562815736 
+Step 77 | loss:1.1006416082382202 lr:1.4568573663685267e-05 tokens_per_second_per_gpu:2950.4712628757084 
+Step 78 | loss:1.1318891048431396 lr:1.4556103659340091e-05 tokens_per_second_per_gpu:3420.308972950097 
+Step 79 | loss:1.073240041732788 lr:1.4543461488125208e-05 tokens_per_second_per_gpu:3292.630999889894 
+Step 80 | loss:0.9821822643280029 lr:1.4530647458505908e-05 tokens_per_second_per_gpu:3845.0406649088623 
+Step 81 | loss:1.031591534614563 lr:1.4517661883140769e-05 tokens_per_second_per_gpu:3597.2292553821944 
+Step 82 | loss:1.0589147806167603 lr:1.4504505078874041e-05 tokens_per_second_per_gpu:3923.875717440058 
+Step 83 | loss:1.0883533954620361 lr:1.449117736672791e-05 tokens_per_second_per_gpu:3328.5666750491487 
+Step 84 | loss:0.9703829288482666 lr:1.4477679071894659e-05 tokens_per_second_per_gpu:4210.9057239346785 
+Step 85 | loss:1.0380264520645142 lr:1.4464010523728745e-05 tokens_per_second_per_gpu:4028.7523308737022 
+Step 86 | loss:1.0424227714538574 lr:1.445017205573875e-05 tokens_per_second_per_gpu:3067.892409170815 
+Step 87 | loss:0.9402433037757874 lr:1.4436164005579258e-05 tokens_per_second_per_gpu:3727.9697325163597 
+Step 88 | loss:0.9366852641105652 lr:1.4421986715042602e-05 tokens_per_second_per_gpu:3359.9819616780637 
+Step 89 | loss:0.9120358824729919 lr:1.4407640530050532e-05 tokens_per_second_per_gpu:3560.9878740376766 
+Step 90 | loss:1.0565634965896606 lr:1.4393125800645775e-05 tokens_per_second_per_gpu:3406.1127910038686 
+Step 91 | loss:1.0152472257614136 lr:1.4378442880983492e-05 tokens_per_second_per_gpu:3213.8003626321415 
+Step 92 | loss:1.1250417232513428 lr:1.4363592129322638e-05 tokens_per_second_per_gpu:3550.8080018136066 
+Step 93 | loss:1.1519802808761597 lr:1.4348573908017218e-05 tokens_per_second_per_gpu:3889.4264438590662 
+Step 94 | loss:1.0634357929229736 lr:1.4333388583507448e-05 tokens_per_second_per_gpu:3799.4362078472236 
+Step 95 | loss:0.9236892461776733 lr:1.4318036526310814e-05 tokens_per_second_per_gpu:3107.0781575328906 
+Step 96 | loss:1.1531763076782227 lr:1.4302518111013029e-05 tokens_per_second_per_gpu:3633.8470364745854 
+Step 97 | loss:1.1617469787597656 lr:1.4286833716258899e-05 tokens_per_second_per_gpu:3394.014913251338 
+Step 98 | loss:1.049908995628357 lr:1.4270983724743077e-05 tokens_per_second_per_gpu:3281.7821637586317 
+Step 99 | loss:1.168889045715332 lr:1.425496852320073e-05 tokens_per_second_per_gpu:3536.842535550907 
+Step 100 | loss:1.029691457748413 lr:1.42387885023981e-05 tokens_per_second_per_gpu:3351.7619303741485 
+Step 101 | loss:0.9969059824943542 lr:1.422244405712297e-05 tokens_per_second_per_gpu:3872.7354730770085 
+Step 102 | loss:1.0703043937683105 lr:1.420593558617504e-05 tokens_per_second_per_gpu:3133.1573043410367 
+Step 103 | loss:0.9668957591056824 lr:1.4189263492356176e-05 tokens_per_second_per_gpu:4133.178776143774 
+Step 104 | loss:1.1153146028518677 lr:1.4172428182460605e-05 tokens_per_second_per_gpu:3279.1325303313006 
+Step 105 | loss:1.0925307273864746 lr:1.4155430067264974e-05 tokens_per_second_per_gpu:3290.367388662595 
+Step 106 | loss:0.9424126148223877 lr:1.4138269561518329e-05 tokens_per_second_per_gpu:3823.0717814425543 
+Step 107 | loss:1.008542537689209 lr:1.4120947083932006e-05 tokens_per_second_per_gpu:3747.7812395683623 
+Step 108 | loss:1.0429927110671997 lr:1.4103463057169398e-05 tokens_per_second_per_gpu:3590.085848047025 
+Step 109 | loss:1.0133060216903687 lr:1.4085817907835657e-05 tokens_per_second_per_gpu:3174.886936324355 
+Step 110 | loss:0.9840688109397888 lr:1.4068012066467276e-05 tokens_per_second_per_gpu:3362.005778821448 
+Step 111 | loss:0.9953323006629944 lr:1.4050045967521587e-05 tokens_per_second_per_gpu:3757.2369577327404 
+Step 112 | loss:1.2065812349319458 lr:1.4031920049366161e-05 tokens_per_second_per_gpu:3325.905289950363 
+Step 113 | loss:0.941290020942688 lr:1.4013634754268107e-05 tokens_per_second_per_gpu:3273.1097453960997 
+Step 114 | loss:1.006861686706543 lr:1.3995190528383292e-05 tokens_per_second_per_gpu:3408.69735024395 
+Step 115 | loss:1.0291639566421509 lr:1.397658782174544e-05 tokens_per_second_per_gpu:3274.3777461502764 
+Step 116 | loss:0.9624848961830139 lr:1.3957827088255166e-05 tokens_per_second_per_gpu:3524.4278903848763 
+Step 117 | loss:1.0975605249404907 lr:1.3938908785668893e-05 tokens_per_second_per_gpu:3103.05657743153 
+Step 118 | loss:1.0946214199066162 lr:1.3919833375587679e-05 tokens_per_second_per_gpu:3842.6176588735884 
+Step 119 | loss:1.0759072303771973 lr:1.3900601323445961e-05 tokens_per_second_per_gpu:3808.512164666663 
+Step 120 | loss:1.01813805103302 lr:1.3881213098500202e-05 tokens_per_second_per_gpu:3999.2491046466607 
+Step 121 | loss:1.0882763862609863 lr:1.3861669173817427e-05 tokens_per_second_per_gpu:3427.5146935052067 
+Step 122 | loss:0.9825055599212646 lr:1.3841970026263695e-05 tokens_per_second_per_gpu:3473.233743555214 
+Step 123 | loss:1.099984884262085 lr:1.382211613649246e-05 tokens_per_second_per_gpu:3717.821725752323 
+Step 124 | loss:1.0696111917495728 lr:1.3802107988932832e-05 tokens_per_second_per_gpu:3082.1936590885393 
+Step 125 | loss:1.1309176683425903 lr:1.3781946071777777e-05 tokens_per_second_per_gpu:3077.2164906951607 
+Step 126 | loss:1.0581340789794922 lr:1.3761630876972183e-05 tokens_per_second_per_gpu:3387.403226559853 
+Step 127 | loss:1.0505608320236206 lr:1.3741162900200874e-05 tokens_per_second_per_gpu:3411.845981322547 
+Step 128 | loss:0.9377124309539795 lr:1.3720542640876514e-05 tokens_per_second_per_gpu:3386.120190777436 
+Step 129 | loss:1.0050909519195557 lr:1.3699770602127406e-05 tokens_per_second_per_gpu:3239.424588950932 
+Step 130 | loss:1.0426461696624756 lr:1.3678847290785237e-05 tokens_per_second_per_gpu:3827.433691106946 
+Step 131 | loss:1.0819664001464844 lr:1.3657773217372694e-05 tokens_per_second_per_gpu:3468.9528124343155 
+Step 132 | loss:0.9391593933105469 lr:1.3636548896091019e-05 tokens_per_second_per_gpu:4109.727828655714 
+Step 133 | loss:1.130295753479004 lr:1.3615174844807451e-05 tokens_per_second_per_gpu:3672.415503816543 
+Step 134 | loss:0.9968859553337097 lr:1.359365158504261e-05 tokens_per_second_per_gpu:4390.05247525661 
+Step 135 | loss:1.1019611358642578 lr:1.3571979641957745e-05 tokens_per_second_per_gpu:3287.8714995416 
+Step 136 | loss:1.047814965248108 lr:1.3550159544341948e-05 tokens_per_second_per_gpu:3286.5090224768937 
+Step 137 | loss:1.041077733039856 lr:1.3528191824599228e-05 tokens_per_second_per_gpu:3717.2716801705315 
+Step 138 | loss:1.0515552759170532 lr:1.3506077018735533e-05 tokens_per_second_per_gpu:3202.3379318776147 
+Step 139 | loss:1.1296720504760742 lr:1.3483815666345674e-05 tokens_per_second_per_gpu:3258.4210318353244 
+Step 140 | loss:1.0627872943878174 lr:1.3461408310600151e-05 tokens_per_second_per_gpu:3194.7948425297013 
+Step 141 | loss:1.0591386556625366 lr:1.3438855498231901e-05 tokens_per_second_per_gpu:3233.8322219652996 
+Step 142 | loss:1.2165062427520752 lr:1.3416157779522969e-05 tokens_per_second_per_gpu:2700.8847148372674 
+Step 143 | loss:1.1000216007232666 lr:1.339331570829106e-05 tokens_per_second_per_gpu:3090.8040651670713 
+Step 144 | loss:1.0945175886154175 lr:1.3370329841876049e-05 tokens_per_second_per_gpu:3016.1521681626327 
+Step 145 | loss:1.0223222970962524 lr:1.3347200741126368e-05 tokens_per_second_per_gpu:3864.9123751265906 
+Step 146 | loss:1.0384180545806885 lr:1.3323928970385318e-05 tokens_per_second_per_gpu:3093.045925373374 
+Step 147 | loss:1.0387237071990967 lr:1.3300515097477319e-05 tokens_per_second_per_gpu:3583.5866748553367 
+Step 148 | loss:1.0544804334640503 lr:1.3276959693694032e-05 tokens_per_second_per_gpu:4013.569528765113 
+Step 149 | loss:1.0835425853729248 lr:1.3253263333780434e-05 tokens_per_second_per_gpu:3747.288931376117 
+Step 150 | loss:1.0735474824905396 lr:1.3229426595920794e-05 tokens_per_second_per_gpu:3016.964949179244 
+Step 151 | loss:1.1732394695281982 lr:1.3205450061724554e-05 tokens_per_second_per_gpu:3552.8302097626038 
+Step 152 | loss:1.0120192766189575 lr:1.3181334316212151e-05 tokens_per_second_per_gpu:3993.2864465929683 
+Step 153 | loss:1.098035216331482 lr:1.3157079947800736e-05 tokens_per_second_per_gpu:3630.096833881631 
+Step 154 | loss:1.0417951345443726 lr:1.313268754828982e-05 tokens_per_second_per_gpu:4128.380187352026 
+Step 155 | loss:1.0801373720169067 lr:1.3108157712846833e-05 tokens_per_second_per_gpu:3170.1141756442244 
+Step 156 | loss:0.9469473361968994 lr:1.3083491039992596e-05 tokens_per_second_per_gpu:3746.1269785447903 
+Step 157 | loss:1.1583002805709839 lr:1.3058688131586727e-05 tokens_per_second_per_gpu:2958.969451708439 
+Step 158 | loss:1.0937081575393677 lr:1.3033749592812955e-05 tokens_per_second_per_gpu:3625.0738232335902 
+Step 159 | loss:1.041576623916626 lr:1.3008676032164346e-05 tokens_per_second_per_gpu:3174.5789362480195 
+Step 160 | loss:1.1431220769882202 lr:1.2983468061428455e-05 tokens_per_second_per_gpu:3186.8973435756893 
+Step 161 | loss:1.060421347618103 lr:1.2958126295672419e-05 tokens_per_second_per_gpu:3658.4590960017413 
+Step 162 | loss:0.9672384262084961 lr:1.293265135322792e-05 tokens_per_second_per_gpu:3234.76727249106 
+Step 163 | loss:1.1202409267425537 lr:1.290704385567612e-05 tokens_per_second_per_gpu:3683.5608176507158 
+Step 164 | loss:1.1316494941711426 lr:1.2881304427832483e-05 tokens_per_second_per_gpu:2893.5620231739244 
+Step 165 | loss:1.0530860424041748 lr:1.2855433697731538e-05 tokens_per_second_per_gpu:3596.0516529135816 
+Step 166 | loss:0.9973039627075195 lr:1.2829432296611547e-05 tokens_per_second_per_gpu:3239.2785653439887 
+Step 167 | loss:1.0723451375961304 lr:1.2803300858899106e-05 tokens_per_second_per_gpu:3442.490376433312 
+Step 168 | loss:1.0164995193481445 lr:1.2777040022193672e-05 tokens_per_second_per_gpu:3164.9946873077406 
+Step 169 | loss:0.9852588176727295 lr:1.275065042725199e-05 tokens_per_second_per_gpu:3323.0232306554753 
+Step 170 | loss:1.0778350830078125 lr:1.2724132717972478e-05 tokens_per_second_per_gpu:2830.1534574906404 
+Step 171 | loss:1.1002942323684692 lr:1.26974875413795e-05 tokens_per_second_per_gpu:3511.467336305257 
+Step 172 | loss:1.0973403453826904 lr:1.267071554760759e-05 tokens_per_second_per_gpu:4210.974619365593 
+Step 173 | loss:1.0474810600280762 lr:1.264381738988558e-05 tokens_per_second_per_gpu:3633.5914296190986 
+Step 174 | loss:1.107753038406372 lr:1.2616793724520665e-05 tokens_per_second_per_gpu:3379.4868101415323 
+Step 175 | loss:1.0912103652954102 lr:1.2589645210882397e-05 tokens_per_second_per_gpu:3268.3413161504523 
+Step 176 | loss:0.9168727397918701 lr:1.256237251138658e-05 tokens_per_second_per_gpu:3584.472835944426 
+Step 177 | loss:1.0452046394348145 lr:1.2534976291479122e-05 tokens_per_second_per_gpu:3418.838663433972 
+Step 178 | loss:1.0404298305511475 lr:1.2507457219619796e-05 tokens_per_second_per_gpu:2913.0762752618957 
+Step 179 | loss:0.9873407483100891 lr:1.247981596726592e-05 tokens_per_second_per_gpu:3476.05126598452 
+Step 180 | loss:0.9315714836120605 lr:1.245205320885598e-05 tokens_per_second_per_gpu:4233.830633953143 
+Step 181 | loss:1.0789451599121094 lr:1.2424169621793182e-05 tokens_per_second_per_gpu:3855.9909233022713 
+Step 182 | loss:1.0372527837753296 lr:1.2396165886428913e-05 tokens_per_second_per_gpu:3188.532121002085 
+Step 183 | loss:0.9855374097824097 lr:1.2368042686046139e-05 tokens_per_second_per_gpu:3360.102286992968 
+Step 184 | loss:0.9893398880958557 lr:1.233980070684274e-05 tokens_per_second_per_gpu:3529.360812306045 
+Step 185 | loss:1.0867488384246826 lr:1.2311440637914766e-05 tokens_per_second_per_gpu:3179.6720985050933 
+Step 186 | loss:1.136749267578125 lr:1.228296317123962e-05 tokens_per_second_per_gpu:3040.5648519621095 
+Step 187 | loss:1.0160714387893677 lr:1.2254369001659178e-05 tokens_per_second_per_gpu:2636.2919865625445 
+Step 188 | loss:0.9412505030632019 lr:1.2225658826862835e-05 tokens_per_second_per_gpu:3556.44720893037 
+Step 189 | loss:1.0352046489715576 lr:1.219683334737047e-05 tokens_per_second_per_gpu:3189.685627683808 
+Step 190 | loss:1.0164215564727783 lr:1.2167893266515368e-05 tokens_per_second_per_gpu:3481.096336571031 
+Step 191 | loss:0.9528836607933044 lr:1.2138839290427062e-05 tokens_per_second_per_gpu:3763.6803007713847 
+Step 192 | loss:0.874556303024292 lr:1.210967212801408e-05 tokens_per_second_per_gpu:3543.1880047946393 
+Step 193 | loss:0.9521118402481079 lr:1.208039249094668e-05 tokens_per_second_per_gpu:3085.3159173416116 
+Step 194 | loss:0.9902434349060059 lr:1.2051001093639451e-05 tokens_per_second_per_gpu:3872.794807892901 
+Step 195 | loss:1.0282405614852905 lr:1.2021498653233912e-05 tokens_per_second_per_gpu:3357.2798676727166 
+Step 196 | loss:1.0458389520645142 lr:1.1991885889581001e-05 tokens_per_second_per_gpu:3481.726431927051 
+Step 197 | loss:1.06132972240448 lr:1.1962163525223505e-05 tokens_per_second_per_gpu:3642.1604094854006 
+Step 198 | loss:1.0513228178024292 lr:1.1932332285378438e-05 tokens_per_second_per_gpu:3643.609473789504 
+Step 199 | loss:1.0699129104614258 lr:1.1902392897919344e-05 tokens_per_second_per_gpu:4025.895506710408 
+Step 200 | loss:1.0457760095596313 lr:1.187234609335854e-05 tokens_per_second_per_gpu:3671.934888630731 
+Step 201 | loss:1.0659315586090088 lr:1.1842192604829286e-05 tokens_per_second_per_gpu:3569.504587981394 
+Step 202 | loss:0.980949878692627 lr:1.1811933168067903e-05 tokens_per_second_per_gpu:3335.3855968235766 
+Step 203 | loss:1.0401633977890015 lr:1.1781568521395815e-05 tokens_per_second_per_gpu:3991.0346971342574 
+Step 204 | loss:1.008753776550293 lr:1.1751099405701535e-05 tokens_per_second_per_gpu:3817.479518764033 
+Step 205 | loss:0.9701691269874573 lr:1.1720526564422593e-05 tokens_per_second_per_gpu:3387.821746371587 
+Step 206 | loss:1.122227430343628 lr:1.1689850743527394e-05 tokens_per_second_per_gpu:3950.7440411904113 
+Step 207 | loss:1.0229696035385132 lr:1.1659072691497014e-05 tokens_per_second_per_gpu:4086.9911967317607 
+Step 208 | loss:1.0376312732696533 lr:1.1628193159306939e-05 tokens_per_second_per_gpu:3460.2927592053616 
+Step 209 | loss:1.0910463333129883 lr:1.1597212900408738e-05 tokens_per_second_per_gpu:3441.665664549203 
+Step 210 | loss:0.951766848564148 lr:1.1566132670711691e-05 tokens_per_second_per_gpu:3765.2848616356405 
+Step 211 | loss:1.1121230125427246 lr:1.1534953228564325e-05 tokens_per_second_per_gpu:3348.8906808140177 
+Step 212 | loss:0.9478145241737366 lr:1.1503675334735933e-05 tokens_per_second_per_gpu:3729.1377760854084 
+Step 213 | loss:1.1243412494659424 lr:1.1472299752397989e-05 tokens_per_second_per_gpu:3466.9044365152718 
+Step 214 | loss:0.9882078170776367 lr:1.1440827247105546e-05 tokens_per_second_per_gpu:3301.1196001394637 
+Step 215 | loss:1.0635075569152832 lr:1.140925858677855e-05 tokens_per_second_per_gpu:3329.520308962072 
+Step 216 | loss:0.9767946004867554 lr:1.1377594541683095e-05 tokens_per_second_per_gpu:3282.602115501923 
+Step 217 | loss:0.9662872552871704 lr:1.134583588441264e-05 tokens_per_second_per_gpu:3304.2082107578244 
+Step 218 | loss:0.9389134049415588 lr:1.1313983389869154e-05 tokens_per_second_per_gpu:3737.964426771077 
+Step 219 | loss:1.0151091814041138 lr:1.1282037835244205e-05 tokens_per_second_per_gpu:3664.737039346888 
+Step 220 | loss:1.0414142608642578 lr:1.125e-05 tokens_per_second_per_gpu:3906.458243931466 
+Step 221 | loss:1.0948292016983032 lr:1.121787066585037e-05 tokens_per_second_per_gpu:3126.7367416868815 
+Step 222 | loss:1.1479017734527588 lr:1.118565061674169e-05 tokens_per_second_per_gpu:3310.893821872108 
+Step 223 | loss:1.1810510158538818 lr:1.1153340638833753e-05 tokens_per_second_per_gpu:3048.5584944239317 
+Step 224 | loss:0.9962694048881531 lr:1.1120941520480588e-05 tokens_per_second_per_gpu:3589.2994193950644 
+Step 225 | loss:1.0925935506820679 lr:1.1088454052211226e-05 tokens_per_second_per_gpu:3637.8874197495197 
+Step 226 | loss:1.0008320808410645 lr:1.1055879026710413e-05 tokens_per_second_per_gpu:3710.3314714150306 
+Step 227 | loss:1.0037486553192139 lr:1.102321723879926e-05 tokens_per_second_per_gpu:3530.8579202116766 
+Step 228 | loss:1.0278898477554321 lr:1.0990469485415859e-05 tokens_per_second_per_gpu:3603.3149217944747 
+Step 229 | loss:0.9470862150192261 lr:1.0957636565595835e-05 tokens_per_second_per_gpu:3314.542239706414 
+Step 230 | loss:0.974503755569458 lr:1.0924719280452849e-05 tokens_per_second_per_gpu:3547.2378512191153 
+Step 231 | loss:0.9583215713500977 lr:1.0891718433159048e-05 tokens_per_second_per_gpu:3523.465479574953 
+Step 232 | loss:1.0647832155227661 lr:1.0858634828925474e-05 tokens_per_second_per_gpu:4079.209517175555 
+Step 233 | loss:1.0859055519104004 lr:1.0825469274982416e-05 tokens_per_second_per_gpu:3748.6882115494077 
+Step 234 | loss:1.0758495330810547 lr:1.0792222580559706e-05 tokens_per_second_per_gpu:3851.188199977994 
+Step 235 | loss:0.9934150576591492 lr:1.0758895556866984e-05 tokens_per_second_per_gpu:3027.8201410044285 
+Step 236 | loss:0.9749125838279724 lr:1.0725489017073905e-05 tokens_per_second_per_gpu:3824.3620745614776 
+Step 237 | loss:1.1250077486038208 lr:1.0692003776290284e-05 tokens_per_second_per_gpu:3812.241805278313 
+Step 238 | loss:0.9836297631263733 lr:1.0658440651546224e-05 tokens_per_second_per_gpu:3754.6913377732058 
+Step 239 | loss:1.0390492677688599 lr:1.0624800461772173e-05 tokens_per_second_per_gpu:3801.826857915905 
+Step 240 | loss:1.04658842086792 lr:1.059108402777894e-05 tokens_per_second_per_gpu:3247.5606533242094 
+Step 241 | loss:1.1221239566802979 lr:1.0557292172237676e-05 tokens_per_second_per_gpu:3813.4732938265543 
+Step 242 | loss:1.150991678237915 lr:1.0523425719659793e-05 tokens_per_second_per_gpu:3428.596138959447 
+Step 243 | loss:1.139402151107788 lr:1.0489485496376844e-05 tokens_per_second_per_gpu:3350.111202236054 
+Step 244 | loss:0.990974485874176 lr:1.0455472330520378e-05 tokens_per_second_per_gpu:3301.0046192724767 
+Step 245 | loss:1.023050308227539 lr:1.0421387052001705e-05 tokens_per_second_per_gpu:2901.330223329429 
+Step 246 | loss:1.0226176977157593 lr:1.0387230492491678e-05 tokens_per_second_per_gpu:3203.0339991749943 
+Step 247 | loss:1.0416032075881958 lr:1.0353003485400378e-05 tokens_per_second_per_gpu:3180.0409616941442 
+Step 248 | loss:0.9852002859115601 lr:1.0318706865856785e-05 tokens_per_second_per_gpu:3998.775527074616 
+Step 249 | loss:1.1657763719558716 lr:1.028434147068841e-05 tokens_per_second_per_gpu:3891.387438322461 
+Step 250 | loss:1.0519105195999146 lr:1.0249908138400862e-05 tokens_per_second_per_gpu:3888.449070526479 
+Step 251 | loss:1.0405876636505127 lr:1.0215407709157396e-05 tokens_per_second_per_gpu:3619.335466976545 
+Step 252 | loss:1.0283851623535156 lr:1.0180841024758419e-05 tokens_per_second_per_gpu:2752.070363393849 
+Step 253 | loss:1.0298212766647339 lr:1.0146208928620938e-05 tokens_per_second_per_gpu:3307.377280772541 
+Step 254 | loss:0.964581310749054 lr:1.0111512265757992e-05 tokens_per_second_per_gpu:3752.103655989273 
+Step 255 | loss:0.9683664441108704 lr:1.0076751882758025e-05 tokens_per_second_per_gpu:4138.001777589129 
+Step 256 | loss:1.0615788698196411 lr:1.0041928627764238e-05 tokens_per_second_per_gpu:3341.61440477967 
+Step 257 | loss:1.0685681104660034 lr:1.0007043350453889e-05 tokens_per_second_per_gpu:3688.378050058521 
+Step 258 | loss:1.0611388683319092 lr:9.972096902017559e-06 tokens_per_second_per_gpu:3797.8084242404275 
+Step 259 | loss:0.961796760559082 lr:9.937090135138392e-06 tokens_per_second_per_gpu:3767.499020527814 
+Step 260 | loss:0.9383830428123474 lr:9.902023903971282e-06 tokens_per_second_per_gpu:4237.725928428407 
+Step 261 | loss:1.0182300806045532 lr:9.866899064122033e-06 tokens_per_second_per_gpu:3204.7581266878838 
+Step 262 | loss:1.0680441856384277 lr:9.831716472626485e-06 tokens_per_second_per_gpu:3623.368447703484 
+Step 263 | loss:1.0305362939834595 lr:9.796476987929601e-06 tokens_per_second_per_gpu:3667.2284929147554 
+Step 264 | loss:1.0003256797790527 lr:9.761181469864523e-06 tokens_per_second_per_gpu:3335.105931326227 
+Step 265 | loss:1.0257227420806885 lr:9.725830779631588e-06 tokens_per_second_per_gpu:4030.3851390190357 
+Step 266 | loss:1.0340684652328491 lr:9.69042577977732e-06 tokens_per_second_per_gpu:3411.0997028354655 
+Step 267 | loss:1.1450355052947998 lr:9.65496733417338e-06 tokens_per_second_per_gpu:3052.7514833967907 
+Step 268 | loss:1.0675625801086426 lr:9.619456307995492e-06 tokens_per_second_per_gpu:3714.1461111976373 
+Step 269 | loss:1.0681331157684326 lr:9.583893567702329e-06 tokens_per_second_per_gpu:3103.583450091087 
+Step 270 | loss:1.0137324333190918 lr:9.548279981014373e-06 tokens_per_second_per_gpu:3423.384868860896 
+Step 271 | loss:1.0491313934326172 lr:9.512616416892749e-06 tokens_per_second_per_gpu:3583.567062020731 
+Step 272 | loss:0.9289360046386719 lr:9.476903745518007e-06 tokens_per_second_per_gpu:3940.2272562878234 
+Step 273 | loss:0.9665488600730896 lr:9.441142838268906e-06 tokens_per_second_per_gpu:2920.460239292336 
+Step 274 | loss:1.0877517461776733 lr:9.405334567701143e-06 tokens_per_second_per_gpu:2961.311684974825 
+Step 275 | loss:1.0493131875991821 lr:9.369479807526072e-06 tokens_per_second_per_gpu:3966.098017081578 
+Step 276 | loss:1.0874686241149902 lr:9.333579432589371e-06 tokens_per_second_per_gpu:3387.0541722877924 
+Step 277 | loss:0.9608097672462463 lr:9.297634318849712e-06 tokens_per_second_per_gpu:4291.32165742434 
+Step 278 | loss:1.095219612121582 lr:9.26164534335738e-06 tokens_per_second_per_gpu:3076.5752421480315 
+Step 279 | loss:1.0561093091964722 lr:9.225613384232867e-06 tokens_per_second_per_gpu:3212.6037289159076 
+Step 280 | loss:1.0817214250564575 lr:9.189539320645461e-06 tokens_per_second_per_gpu:3659.336877166898 
+Step 281 | loss:1.0177308320999146 lr:9.15342403279179e-06 tokens_per_second_per_gpu:3738.1776707794156 
+Step 282 | loss:0.9710911512374878 lr:9.117268401874329e-06 tokens_per_second_per_gpu:3382.0765709925126 
+Step 283 | loss:1.1531765460968018 lr:9.081073310079919e-06 tokens_per_second_per_gpu:3141.8432625191035 
+Step 284 | loss:1.0935015678405762 lr:9.044839640558238e-06 tokens_per_second_per_gpu:3429.2429675320373 
+Step 285 | loss:0.9385885000228882 lr:9.008568277400246e-06 tokens_per_second_per_gpu:3346.592745218261 
+Step 286 | loss:0.9143052101135254 lr:8.972260105616615e-06 tokens_per_second_per_gpu:3597.7170701513446 
+Step 287 | loss:1.1004893779754639 lr:8.935916011116141e-06 tokens_per_second_per_gpu:3088.5027102851095 
+Step 288 | loss:1.0081392526626587 lr:8.899536880684118e-06 tokens_per_second_per_gpu:3565.6117227664477 
+Step 289 | loss:1.0537937879562378 lr:8.863123601960713e-06 tokens_per_second_per_gpu:3896.8268798437653 
+Step 290 | loss:1.0108102560043335 lr:8.826677063419297e-06 tokens_per_second_per_gpu:2986.3032089900526 
+Step 291 | loss:0.9827962517738342 lr:8.790198154344774e-06 tokens_per_second_per_gpu:3411.9397148980493 
+Step 292 | loss:0.9616813659667969 lr:8.753687764811874e-06 tokens_per_second_per_gpu:4128.635119314705 
+Step 293 | loss:1.0986084938049316 lr:8.717146785663451e-06 tokens_per_second_per_gpu:3044.1432990396206 
+Step 294 | loss:1.010633111000061 lr:8.680576108488722e-06 tokens_per_second_per_gpu:4016.5995836897896 
+Step 295 | loss:1.1678359508514404 lr:8.643976625601543e-06 tokens_per_second_per_gpu:3484.066437413288 
+Step 296 | loss:0.9672271609306335 lr:8.60734923001861e-06 tokens_per_second_per_gpu:3722.9078388086114 
+Step 297 | loss:1.0644941329956055 lr:8.570694815437684e-06 tokens_per_second_per_gpu:3674.078084922676 
+Step 298 | loss:1.0218087434768677 lr:8.534014276215784e-06 tokens_per_second_per_gpu:3719.1251320960077 
+Step 299 | loss:0.8616893291473389 lr:8.497308507347358e-06 tokens_per_second_per_gpu:3980.0883290351844 
+Step 300 | loss:1.200434923171997 lr:8.460578404442452e-06 tokens_per_second_per_gpu:3537.9024408752107 
+Step 301 | loss:1.2012302875518799 lr:8.423824863704859e-06 tokens_per_second_per_gpu:3341.0804733107207 
+Step 302 | loss:1.0611824989318848 lr:8.387048781910243e-06 tokens_per_second_per_gpu:3399.3160464558623 
+Step 303 | loss:1.0133317708969116 lr:8.350251056384267e-06 tokens_per_second_per_gpu:3396.784053005488 
+Step 304 | loss:1.0075584650039673 lr:8.313432584980693e-06 tokens_per_second_per_gpu:3170.834006045349 
+Step 305 | loss:1.0057969093322754 lr:8.276594266059473e-06 tokens_per_second_per_gpu:3349.7058914501686 
+Step 306 | loss:0.9875553846359253 lr:8.239736998464839e-06 tokens_per_second_per_gpu:3283.555251517354 
+Step 307 | loss:1.0624175071716309 lr:8.202861681503362e-06 tokens_per_second_per_gpu:3288.3057635972655 
+Step 308 | loss:1.0439351797103882 lr:8.165969214922011e-06 tokens_per_second_per_gpu:3257.713880411403 
+Step 309 | loss:1.021206021308899 lr:8.129060498886204e-06 tokens_per_second_per_gpu:3443.1099672423125 
+Step 310 | loss:1.0511139631271362 lr:8.09213643395784e-06 tokens_per_second_per_gpu:3937.8803984424717 
+Step 311 | loss:0.9559040069580078 lr:8.05519792107332e-06 tokens_per_second_per_gpu:3379.1804990828364 
+Step 312 | loss:1.048850655555725 lr:8.018245861521585e-06 tokens_per_second_per_gpu:3090.2532971700753 
+Step 313 | loss:1.1831482648849487 lr:7.981281156922097e-06 tokens_per_second_per_gpu:3205.4884774853103 
+Step 314 | loss:1.065642237663269 lr:7.944304709202857e-06 tokens_per_second_per_gpu:3615.7381651115775 
+Step 315 | loss:0.9639590978622437 lr:7.9073174205784e-06 tokens_per_second_per_gpu:3895.6258291409545 
+Step 316 | loss:1.0430667400360107 lr:7.870320193527773e-06 tokens_per_second_per_gpu:3228.924479386842 
+Step 317 | loss:1.104460597038269 lr:7.833313930772514e-06 tokens_per_second_per_gpu:3807.3283800568493 
+Step 318 | loss:1.0125938653945923 lr:7.796299535254633e-06 tokens_per_second_per_gpu:3032.071204987172 
+Step 319 | loss:1.064937949180603 lr:7.759277910114582e-06 tokens_per_second_per_gpu:3479.6920455124077 
+Step 320 | loss:1.0094972848892212 lr:7.722249958669199e-06 tokens_per_second_per_gpu:3269.3602673037053 
+Step 321 | loss:1.0703675746917725 lr:7.685216584389697e-06 tokens_per_second_per_gpu:3704.300080154252 
+Step 322 | loss:0.9877763390541077 lr:7.648178690879598e-06 tokens_per_second_per_gpu:3353.1459094211864 
+Step 323 | loss:1.12434983253479 lr:7.611137181852695e-06 tokens_per_second_per_gpu:3122.7128974218263 
+Step 324 | loss:1.0240429639816284 lr:7.574092961110993e-06 tokens_per_second_per_gpu:3098.34410006354 
+Step 325 | loss:0.9952061176300049 lr:7.537046932522668e-06 tokens_per_second_per_gpu:3604.4779903256135 
+Step 326 | loss:1.011582612991333 lr:7.5e-06 tokens_per_second_per_gpu:3283.509048393859 
+Step 327 | loss:0.9831445217132568 lr:7.462953067477332e-06 tokens_per_second_per_gpu:3775.59490266086 
+Step 328 | loss:1.1675689220428467 lr:7.425907038889008e-06 tokens_per_second_per_gpu:3642.4454328719376 
+Step 329 | loss:1.0117154121398926 lr:7.388862818147305e-06 tokens_per_second_per_gpu:3161.511403244572 
+Step 330 | loss:0.9967679381370544 lr:7.351821309120403e-06 tokens_per_second_per_gpu:4076.6176819879342 
+Step 331 | loss:1.0238803625106812 lr:7.314783415610303e-06 tokens_per_second_per_gpu:3682.85016942687 
+Step 332 | loss:1.1032280921936035 lr:7.2777500413308015e-06 tokens_per_second_per_gpu:3018.3267646978616 
+Step 333 | loss:1.0388715267181396 lr:7.240722089885421e-06 tokens_per_second_per_gpu:3664.4967840336526 
+Step 334 | loss:1.0032538175582886 lr:7.203700464745366e-06 tokens_per_second_per_gpu:3878.0272534442006 
+Step 335 | loss:0.9861161112785339 lr:7.166686069227486e-06 tokens_per_second_per_gpu:3704.921578499297 
+Step 336 | loss:1.0723155736923218 lr:7.129679806472228e-06 tokens_per_second_per_gpu:3379.4966782090632 
+Step 337 | loss:1.0907748937606812 lr:7.092682579421598e-06 tokens_per_second_per_gpu:3056.886775193357 
+Step 338 | loss:1.0879602432250977 lr:7.055695290797143e-06 tokens_per_second_per_gpu:3510.6596962005956 
+Step 339 | loss:1.2004570960998535 lr:7.018718843077904e-06 tokens_per_second_per_gpu:2698.696907414987 
+Step 340 | loss:1.1619454622268677 lr:6.981754138478416e-06 tokens_per_second_per_gpu:3246.2572334355586 
+Step 341 | loss:1.1515110731124878 lr:6.944802078926679e-06 tokens_per_second_per_gpu:3833.456255283371 
+Step 342 | loss:0.9102157950401306 lr:6.907863566042161e-06 tokens_per_second_per_gpu:3861.3976159065483 
+Step 343 | loss:1.0276517868041992 lr:6.870939501113796e-06 tokens_per_second_per_gpu:3515.3211579047493 
+Step 344 | loss:1.0000083446502686 lr:6.834030785077989e-06 tokens_per_second_per_gpu:3557.346531908244 
+Step 345 | loss:1.0519853830337524 lr:6.797138318496637e-06 tokens_per_second_per_gpu:3382.344983091482 
+Step 346 | loss:1.0421191453933716 lr:6.7602630015351624e-06 tokens_per_second_per_gpu:3723.6117480414146 
+Step 347 | loss:0.9147570133209229 lr:6.723405733940528e-06 tokens_per_second_per_gpu:4252.460083108924 
+Step 348 | loss:0.9446225166320801 lr:6.68656741501931e-06 tokens_per_second_per_gpu:3618.6484563348768 
+Step 349 | loss:0.9708753228187561 lr:6.649748943615732e-06 tokens_per_second_per_gpu:3283.5603120185374 
+Step 350 | loss:0.989623486995697 lr:6.6129512180897574e-06 tokens_per_second_per_gpu:3714.9557633983477 
+Step 351 | loss:1.0894560813903809 lr:6.57617513629514e-06 tokens_per_second_per_gpu:3633.0728284201673 
+Step 352 | loss:1.115202784538269 lr:6.539421595557549e-06 tokens_per_second_per_gpu:3465.6183875539423 
+Step 353 | loss:0.8921554088592529 lr:6.502691492652643e-06 tokens_per_second_per_gpu:3898.581389798211 
+Step 354 | loss:1.0538413524627686 lr:6.465985723784218e-06 tokens_per_second_per_gpu:3875.7674647016806 
+Step 355 | loss:1.1229345798492432 lr:6.429305184562315e-06 tokens_per_second_per_gpu:3747.439269311531 
+Step 356 | loss:1.0564650297164917 lr:6.392650769981392e-06 tokens_per_second_per_gpu:3671.9056611937544 
+Step 357 | loss:1.0135570764541626 lr:6.356023374398456e-06 tokens_per_second_per_gpu:2899.364909402783 
+Step 358 | loss:1.0615811347961426 lr:6.319423891511278e-06 tokens_per_second_per_gpu:3801.336525876768 
+Step 359 | loss:0.9940143823623657 lr:6.28285321433655e-06 tokens_per_second_per_gpu:3759.510249531142 
+Step 360 | loss:1.051573395729065 lr:6.246312235188126e-06 tokens_per_second_per_gpu:3262.2914203901505 
+Step 361 | loss:1.0494478940963745 lr:6.209801845655227e-06 tokens_per_second_per_gpu:3231.54016416701 
+Step 362 | loss:1.134334921836853 lr:6.173322936580705e-06 tokens_per_second_per_gpu:3571.6658588531845 
+Step 363 | loss:1.132298231124878 lr:6.136876398039287e-06 tokens_per_second_per_gpu:3057.396332653723 
+Step 364 | loss:1.0008654594421387 lr:6.100463119315882e-06 tokens_per_second_per_gpu:3834.101013732286 
+Step 365 | loss:1.1123298406600952 lr:6.0640839888838594e-06 tokens_per_second_per_gpu:3247.3705233572437 
+Step 366 | loss:1.1211216449737549 lr:6.027739894383387e-06 tokens_per_second_per_gpu:2845.3333047002448 
+Step 367 | loss:1.0459622144699097 lr:5.991431722599755e-06 tokens_per_second_per_gpu:3457.97226490014 
+Step 368 | loss:1.0909711122512817 lr:5.955160359441763e-06 tokens_per_second_per_gpu:3221.2013455847036 
+Step 369 | loss:1.1640664339065552 lr:5.918926689920081e-06 tokens_per_second_per_gpu:3451.3559735588115 
+Step 370 | loss:1.2360296249389648 lr:5.882731598125674e-06 tokens_per_second_per_gpu:3289.3973935577333 
+Step 371 | loss:1.0348141193389893 lr:5.846575967208211e-06 tokens_per_second_per_gpu:3435.4094495338736 
+Step 372 | loss:0.9567263722419739 lr:5.810460679354538e-06 tokens_per_second_per_gpu:3576.81900822866 
+Step 373 | loss:0.9700967073440552 lr:5.774386615767134e-06 tokens_per_second_per_gpu:4121.726341513599 
+Step 374 | loss:1.0645740032196045 lr:5.738354656642623e-06 tokens_per_second_per_gpu:3347.2283933756694 
+Step 375 | loss:1.0295037031173706 lr:5.702365681150289e-06 tokens_per_second_per_gpu:3868.2590545517114 
+Step 376 | loss:1.0450406074523926 lr:5.66642056741063e-06 tokens_per_second_per_gpu:3477.8887834676907 
+Step 377 | loss:1.0922298431396484 lr:5.630520192473929e-06 tokens_per_second_per_gpu:3405.0405996273425 
+Step 378 | loss:0.9819583296775818 lr:5.594665432298858e-06 tokens_per_second_per_gpu:3530.057502597269 
+Step 379 | loss:1.0701051950454712 lr:5.558857161731094e-06 tokens_per_second_per_gpu:3565.327286781416 
+Step 380 | loss:0.9559437036514282 lr:5.523096254481995e-06 tokens_per_second_per_gpu:3365.6710550463654 
+Step 381 | loss:1.0547300577163696 lr:5.48738358310725e-06 tokens_per_second_per_gpu:3346.3530578743994 
+Step 382 | loss:0.9696499705314636 lr:5.451720018985626e-06 tokens_per_second_per_gpu:3329.928998826801 
+Step 383 | loss:1.0064926147460938 lr:5.416106432297671e-06 tokens_per_second_per_gpu:4094.6995868513072 
+Step 384 | loss:1.0114684104919434 lr:5.380543692004509e-06 tokens_per_second_per_gpu:3475.432549582163 
+Step 385 | loss:0.9692176580429077 lr:5.345032665826621e-06 tokens_per_second_per_gpu:3244.483451213904 
+Step 386 | loss:1.0306236743927002 lr:5.3095742202226825e-06 tokens_per_second_per_gpu:3269.062813082469 
+Step 387 | loss:0.9608441591262817 lr:5.274169220368412e-06 tokens_per_second_per_gpu:3293.0322861856876 
+Step 388 | loss:1.0451374053955078 lr:5.238818530135479e-06 tokens_per_second_per_gpu:2938.7446792343676 
+Step 389 | loss:1.1184594631195068 lr:5.203523012070398e-06 tokens_per_second_per_gpu:4074.2637958572395 
+Step 390 | loss:1.0143616199493408 lr:5.168283527373516e-06 tokens_per_second_per_gpu:3757.202601042462 
+Step 391 | loss:1.0019251108169556 lr:5.133100935877967e-06 tokens_per_second_per_gpu:3643.925463722737 
+Step 392 | loss:0.9685758352279663 lr:5.097976096028719e-06 tokens_per_second_per_gpu:3100.3578523101673 
+Step 393 | loss:1.0199055671691895 lr:5.0629098648616075e-06 tokens_per_second_per_gpu:3384.275878283242 
+Step 394 | loss:1.178783893585205 lr:5.027903097982441e-06 tokens_per_second_per_gpu:3391.4302131315617 
+Step 395 | loss:1.0473383665084839 lr:4.992956649546113e-06 tokens_per_second_per_gpu:3063.8065410727527 
+Step 396 | loss:1.0996811389923096 lr:4.958071372235763e-06 tokens_per_second_per_gpu:3498.2229002983877 
+Step 397 | loss:1.0462857484817505 lr:4.923248117241975e-06 tokens_per_second_per_gpu:4333.282035445803 
+Step 398 | loss:0.9347184300422668 lr:4.8884877342420105e-06 tokens_per_second_per_gpu:3622.574379253927 
+Step 399 | loss:0.9804700613021851 lr:4.853791071379062e-06 tokens_per_second_per_gpu:3226.8869428343696 
+Step 400 | loss:1.1005946397781372 lr:4.8191589752415834e-06 tokens_per_second_per_gpu:3372.8523923167445 
+Step 401 | loss:0.8959193229675293 lr:4.7845922908426035e-06 tokens_per_second_per_gpu:3320.152228865833 
+Step 402 | loss:1.021907925605774 lr:4.75009186159914e-06 tokens_per_second_per_gpu:3777.003290844435 
+Step 403 | loss:1.0068416595458984 lr:4.71565852931159e-06 tokens_per_second_per_gpu:3986.641799795482 
+Step 404 | loss:1.000541090965271 lr:4.681293134143216e-06 tokens_per_second_per_gpu:3747.1434322920663 
+Step 405 | loss:1.2205549478530884 lr:4.646996514599623e-06 tokens_per_second_per_gpu:3820.4594798567286 
+Step 406 | loss:0.9515297412872314 lr:4.612769507508324e-06 tokens_per_second_per_gpu:3795.377292339278 
+Step 407 | loss:1.0238933563232422 lr:4.5786129479982945e-06 tokens_per_second_per_gpu:3229.7823570944547 
+Step 408 | loss:1.0509495735168457 lr:4.544527669479625e-06 tokens_per_second_per_gpu:3349.333952642161 
+Step 409 | loss:1.0613925457000732 lr:4.510514503623155e-06 tokens_per_second_per_gpu:3156.1717446768876 
+Step 410 | loss:1.033313512802124 lr:4.4765742803402094e-06 tokens_per_second_per_gpu:3336.8989972246827 
+Step 411 | loss:1.0496494770050049 lr:4.442707827762323e-06 tokens_per_second_per_gpu:3590.429787817092 
+Step 412 | loss:1.0385499000549316 lr:4.4089159722210605e-06 tokens_per_second_per_gpu:3170.2776264525723 
+Step 413 | loss:0.9418953061103821 lr:4.375199538227827e-06 tokens_per_second_per_gpu:3534.558502200195 
+Step 414 | loss:1.047280192375183 lr:4.341559348453779e-06 tokens_per_second_per_gpu:2733.5207695529916 
+Step 415 | loss:1.022471308708191 lr:4.307996223709717e-06 tokens_per_second_per_gpu:3890.7734867828663 
+Step 416 | loss:1.0086355209350586 lr:4.274510982926099e-06 tokens_per_second_per_gpu:3987.506565566676 
+Step 417 | loss:1.0402534008026123 lr:4.2411044431330156e-06 tokens_per_second_per_gpu:3236.307481873333 
+Step 418 | loss:0.8711655139923096 lr:4.207777419440298e-06 tokens_per_second_per_gpu:3975.2863975021874 
+Step 419 | loss:1.023214340209961 lr:4.174530725017586e-06 tokens_per_second_per_gpu:3554.0476686396737 
+Step 420 | loss:0.9571332335472107 lr:4.141365171074529e-06 tokens_per_second_per_gpu:3525.9512956270514 
+Step 421 | loss:1.0896693468093872 lr:4.108281566840953e-06 tokens_per_second_per_gpu:3153.098645463316 
+Step 422 | loss:0.9747520685195923 lr:4.075280719547152e-06 tokens_per_second_per_gpu:3966.0455004943665 
+Step 423 | loss:1.1110624074935913 lr:4.042363434404165e-06 tokens_per_second_per_gpu:3779.0025960246794 
+Step 424 | loss:1.0604115724563599 lr:4.009530514584142e-06 tokens_per_second_per_gpu:3830.9188953316843 
+Step 425 | loss:1.0563576221466064 lr:3.976782761200741e-06 tokens_per_second_per_gpu:3582.450000860484 
+Step 426 | loss:1.0671446323394775 lr:3.944120973289589e-06 tokens_per_second_per_gpu:2908.494054782393 
+Step 427 | loss:0.9762901067733765 lr:3.911545947788775e-06 tokens_per_second_per_gpu:3053.4349628908967 
+Step 428 | loss:0.9776210188865662 lr:3.879058479519415e-06 tokens_per_second_per_gpu:3544.2355797599025 
+Step 429 | loss:1.0141849517822266 lr:3.846659361166249e-06 tokens_per_second_per_gpu:3297.786365076082 
+Step 430 | loss:1.1040488481521606 lr:3.8143493832583126e-06 tokens_per_second_per_gpu:3338.843244763157 
+Step 431 | loss:1.0626472234725952 lr:3.7821293341496314e-06 tokens_per_second_per_gpu:4418.329163218037 
+Step 432 | loss:1.0949605703353882 lr:3.750000000000002e-06 tokens_per_second_per_gpu:3047.106222452544 
+Step 433 | loss:1.0159730911254883 lr:3.717962164755795e-06 tokens_per_second_per_gpu:3602.6745614483334 
+Step 434 | loss:1.0500991344451904 lr:3.686016610130848e-06 tokens_per_second_per_gpu:3986.7451576661006 
+Step 435 | loss:1.0814403295516968 lr:3.654164115587359e-06 tokens_per_second_per_gpu:3571.832029398707 
+Step 436 | loss:1.0364539623260498 lr:3.622405458316908e-06 tokens_per_second_per_gpu:3642.0438527830474 
+Step 437 | loss:1.1901767253875732 lr:3.5907414132214504e-06 tokens_per_second_per_gpu:3176.521239313863 
+Step 438 | loss:0.9588651657104492 lr:3.5591727528944566e-06 tokens_per_second_per_gpu:3918.860748590204 
+Step 439 | loss:1.0434871912002563 lr:3.5277002476020124e-06 tokens_per_second_per_gpu:3299.17808549516 
+Step 440 | loss:1.0512974262237549 lr:3.496324665264073e-06 tokens_per_second_per_gpu:3635.173332567157 
+Step 441 | loss:1.231705665588379 lr:3.465046771435676e-06 tokens_per_second_per_gpu:3074.205941042435 
+Step 442 | loss:1.0789237022399902 lr:3.4338673292883108e-06 tokens_per_second_per_gpu:3693.2828587088543 
+Step 443 | loss:1.2193183898925781 lr:3.4027870995912626e-06 tokens_per_second_per_gpu:3431.2840987432733 
+Step 444 | loss:1.0715184211730957 lr:3.3718068406930634e-06 tokens_per_second_per_gpu:3682.480389381471 
+Step 445 | loss:1.0878655910491943 lr:3.3409273085029877e-06 tokens_per_second_per_gpu:3247.4307989923645 
+Step 446 | loss:1.1202605962753296 lr:3.3101492564726074e-06 tokens_per_second_per_gpu:2868.354735858521 
+Step 447 | loss:0.9288511276245117 lr:3.279473435577409e-06 tokens_per_second_per_gpu:3563.062402935358 
+Step 448 | loss:1.0990906953811646 lr:3.2489005942984675e-06 tokens_per_second_per_gpu:2858.893387808962 
+Step 449 | loss:0.9979101419448853 lr:3.218431478604187e-06 tokens_per_second_per_gpu:3678.9613236315463 
+Step 450 | loss:1.1519663333892822 lr:3.188066831932098e-06 tokens_per_second_per_gpu:3727.64506347197 
+Step 451 | loss:1.0830491781234741 lr:3.157807395170714e-06 tokens_per_second_per_gpu:3000.9712833795475 
+Step 452 | loss:0.953801691532135 lr:3.127653906641461e-06 tokens_per_second_per_gpu:3259.1973993559814 
+Step 453 | loss:1.0369517803192139 lr:3.097607102080657e-06 tokens_per_second_per_gpu:3511.980574010068 
+Step 454 | loss:1.092458724975586 lr:3.067667714621564e-06 tokens_per_second_per_gpu:3294.265502138293 
+Step 455 | loss:0.936871349811554 lr:3.037836474776495e-06 tokens_per_second_per_gpu:3481.5948657632803 
+Step 456 | loss:1.1124027967453003 lr:3.008114110419e-06 tokens_per_second_per_gpu:3812.8193336820773 
+Step 457 | loss:1.0468907356262207 lr:2.978501346766086e-06 tokens_per_second_per_gpu:3151.5910084710267 
+Step 458 | loss:1.0531938076019287 lr:2.948998906360552e-06 tokens_per_second_per_gpu:4215.472517871868 
+Step 459 | loss:1.041715145111084 lr:2.9196075090533224e-06 tokens_per_second_per_gpu:3710.2677325476725 
+Step 460 | loss:1.0383410453796387 lr:2.890327871985922e-06 tokens_per_second_per_gpu:3652.23350104369 
+Step 461 | loss:1.0480257272720337 lr:2.8611607095729393e-06 tokens_per_second_per_gpu:3064.9383393059015 
+Step 462 | loss:0.9538127779960632 lr:2.8321067334846317e-06 tokens_per_second_per_gpu:3769.898140676965 
+Step 463 | loss:0.9850819110870361 lr:2.8031666526295325e-06 tokens_per_second_per_gpu:3406.614961871896 
+Step 464 | loss:1.0228904485702515 lr:2.7743411731371677e-06 tokens_per_second_per_gpu:3569.9039916901243 
+Step 465 | loss:0.9739654064178467 lr:2.7456309983408215e-06 tokens_per_second_per_gpu:3855.6404996182887 
+Step 466 | loss:1.0107979774475098 lr:2.7170368287603812e-06 tokens_per_second_per_gpu:2971.2864568059426 
+Step 467 | loss:1.092708945274353 lr:2.6885593620852362e-06 tokens_per_second_per_gpu:2782.0899935873463 
+Step 468 | loss:1.1168324947357178 lr:2.660199293157263e-06 tokens_per_second_per_gpu:3767.413743825975 
+Step 469 | loss:1.0790526866912842 lr:2.6319573139538637e-06 tokens_per_second_per_gpu:3453.5127524794507 
+Step 470 | loss:1.0022902488708496 lr:2.603834113571089e-06 tokens_per_second_per_gpu:3047.035744052629 
+Step 471 | loss:0.9685212969779968 lr:2.575830378206819e-06 tokens_per_second_per_gpu:3106.6813389813046 
+Step 472 | loss:0.9829853773117065 lr:2.547946791144022e-06 tokens_per_second_per_gpu:3517.443534427285 
+Step 473 | loss:1.0006290674209595 lr:2.520184032734084e-06 tokens_per_second_per_gpu:3299.938446528448 
+Step 474 | loss:1.06125009059906 lr:2.492542780380207e-06 tokens_per_second_per_gpu:3319.6900284288845 
+Step 475 | loss:1.1993439197540283 lr:2.4650237085208767e-06 tokens_per_second_per_gpu:3576.5268885175046 
+Step 476 | loss:0.9784356951713562 lr:2.4376274886134225e-06 tokens_per_second_per_gpu:3467.613615187959 
+Step 477 | loss:0.9801285862922668 lr:2.4103547891176042e-06 tokens_per_second_per_gpu:4156.110968467437 
+Step 478 | loss:1.2105839252471924 lr:2.3832062754793375e-06 tokens_per_second_per_gpu:3410.3132050226804 
+Step 479 | loss:1.0551785230636597 lr:2.356182610114421e-06 tokens_per_second_per_gpu:3536.043611031053 
+Step 480 | loss:1.0802967548370361 lr:2.3292844523924132e-06 tokens_per_second_per_gpu:3609.396572236827 
+Step 481 | loss:1.0849721431732178 lr:2.3025124586205e-06 tokens_per_second_per_gpu:3036.075258849583 
+Step 482 | loss:1.0422958135604858 lr:2.2758672820275245e-06 tokens_per_second_per_gpu:3702.107907772152 
+Step 483 | loss:1.0579990148544312 lr:2.2493495727480106e-06 tokens_per_second_per_gpu:3823.945807707913 
+Step 484 | loss:1.0904960632324219 lr:2.22295997780633e-06 tokens_per_second_per_gpu:3312.442269740297 
+Step 485 | loss:1.0701911449432373 lr:2.196699141100894e-06 tokens_per_second_per_gpu:3242.470040399053 
+Step 486 | loss:0.9568579196929932 lr:2.170567703388454e-06 tokens_per_second_per_gpu:3632.613290114698 
+Step 487 | loss:1.0242254734039307 lr:2.1445663022684626e-06 tokens_per_second_per_gpu:3655.3842931955455 
+Step 488 | loss:1.0032799243927002 lr:2.1186955721675145e-06 tokens_per_second_per_gpu:3707.926602365104 
+Step 489 | loss:1.0453135967254639 lr:2.092956144323881e-06 tokens_per_second_per_gpu:2869.137361469829 
+Step 490 | loss:1.2181352376937866 lr:2.067348646772079e-06 tokens_per_second_per_gpu:3293.3080266736984 
+Step 491 | loss:1.000616192817688 lr:2.041873704327583e-06 tokens_per_second_per_gpu:3715.517462012418 
+Step 492 | loss:1.0542888641357422 lr:2.0165319385715436e-06 tokens_per_second_per_gpu:3830.4390289474027 
+Step 493 | loss:1.0258901119232178 lr:1.991323967835658e-06 tokens_per_second_per_gpu:3028.369221973012 
+Step 494 | loss:0.9300064444541931 lr:1.966250407187045e-06 tokens_per_second_per_gpu:3930.731519264302 
+Step 495 | loss:1.0918774604797363 lr:1.9413118684132744e-06 tokens_per_second_per_gpu:3462.5265384350296 
+Step 496 | loss:1.0639344453811646 lr:1.916508960007404e-06 tokens_per_second_per_gpu:3684.844309448488 
+Step 497 | loss:1.0016651153564453 lr:1.8918422871531677e-06 tokens_per_second_per_gpu:3102.599497609656 
+Step 498 | loss:1.0390342473983765 lr:1.8673124517101783e-06 tokens_per_second_per_gpu:3670.164801395993 
+Step 499 | loss:1.210891842842102 lr:1.842920052199263e-06 tokens_per_second_per_gpu:2854.611146514477 
+Step 500 | loss:1.1124374866485596 lr:1.818665683787849e-06 tokens_per_second_per_gpu:3270.6689217420526 
+Step 501 | loss:0.9718752503395081 lr:1.794549938275447e-06 tokens_per_second_per_gpu:4170.2565142940675 
+Step 502 | loss:1.1475387811660767 lr:1.7705734040792066e-06 tokens_per_second_per_gpu:3874.4289722672997 
+Step 503 | loss:0.9939885139465332 lr:1.7467366662195653e-06 tokens_per_second_per_gpu:4018.9157193209103 
+Step 504 | loss:1.060147762298584 lr:1.7230403063059688e-06 tokens_per_second_per_gpu:3884.5121540563578 
+Step 505 | loss:1.153907060623169 lr:1.6994849025226819e-06 tokens_per_second_per_gpu:3476.133514653969 
+Step 506 | loss:0.9286587238311768 lr:1.6760710296146827e-06 tokens_per_second_per_gpu:3320.296836386092 
+Step 507 | loss:1.1152493953704834 lr:1.6527992588736346e-06 tokens_per_second_per_gpu:2898.2952105804293 
+Step 508 | loss:1.1163067817687988 lr:1.6296701581239514e-06 tokens_per_second_per_gpu:3230.075370963527 
+Step 509 | loss:1.0790047645568848 lr:1.6066842917089406e-06 tokens_per_second_per_gpu:3259.4500583104896 
+Step 510 | loss:0.9519661068916321 lr:1.5838422204770304e-06 tokens_per_second_per_gpu:3670.231481091447 
+Step 511 | loss:0.992994487285614 lr:1.5611445017680991e-06 tokens_per_second_per_gpu:3799.1403263844904 
+Step 512 | loss:1.111997127532959 lr:1.5385916893998497e-06 tokens_per_second_per_gpu:3600.044046651026 
+Step 513 | loss:0.9813446998596191 lr:1.5161843336543285e-06 tokens_per_second_per_gpu:3486.542256282655 
+Step 514 | loss:1.058476209640503 lr:1.4939229812644679e-06 tokens_per_second_per_gpu:3923.009117188916 
+Step 515 | loss:0.9895375967025757 lr:1.4718081754007753e-06 tokens_per_second_per_gpu:3754.8223465225924 
+Step 516 | loss:1.079698085784912 lr:1.4498404556580525e-06 tokens_per_second_per_gpu:3386.562332279495 
+Step 517 | loss:1.0072911977767944 lr:1.4280203580422537e-06 tokens_per_second_per_gpu:3733.3749612885863 
+Step 518 | loss:1.0473989248275757 lr:1.4063484149573902e-06 tokens_per_second_per_gpu:3459.389528812242 
+Step 519 | loss:0.9953117370605469 lr:1.3848251551925475e-06 tokens_per_second_per_gpu:3253.9885230172995 
+Step 520 | loss:1.0296379327774048 lr:1.3634511039089819e-06 tokens_per_second_per_gpu:3249.8868818068645 
+Step 521 | loss:1.0001991987228394 lr:1.3422267826273052e-06 tokens_per_second_per_gpu:3295.052521167191 
+Step 522 | loss:1.0914515256881714 lr:1.321152709214763e-06 tokens_per_second_per_gpu:3418.445922794606 
+Step 523 | loss:1.0070934295654297 lr:1.3002293978725935e-06 tokens_per_second_per_gpu:3468.4665497148526 
+Step 524 | loss:0.991290807723999 lr:1.2794573591234868e-06 tokens_per_second_per_gpu:3655.7877168779905 
+Step 525 | loss:0.9848833680152893 lr:1.2588370997991262e-06 tokens_per_second_per_gpu:3221.000627253079 
+Step 526 | loss:1.0280349254608154 lr:1.2383691230278197e-06 tokens_per_second_per_gpu:3589.7060395621884 
+Step 527 | loss:1.0546156167984009 lr:1.2180539282222252e-06 tokens_per_second_per_gpu:3325.8819649807083 
+Step 528 | loss:0.9586650729179382 lr:1.1978920110671688e-06 tokens_per_second_per_gpu:3187.053627556894 
+Step 529 | loss:1.0085161924362183 lr:1.1778838635075415e-06 tokens_per_second_per_gpu:3512.8972208771875 
+Step 530 | loss:0.9839969873428345 lr:1.1580299737363037e-06 tokens_per_second_per_gpu:3345.8963367573137 
+Step 531 | loss:1.0338804721832275 lr:1.1383308261825748e-06 tokens_per_second_per_gpu:3539.808114559959 
+Step 532 | loss:1.1831824779510498 lr:1.1187869014997992e-06 tokens_per_second_per_gpu:3428.417581753952 
+Step 533 | loss:1.0245225429534912 lr:1.0993986765540403e-06 tokens_per_second_per_gpu:3259.282719856141 
+Step 534 | loss:1.022165060043335 lr:1.080166624412322e-06 tokens_per_second_per_gpu:3233.591194250682 
+Step 535 | loss:1.1499828100204468 lr:1.0610912143311096e-06 tokens_per_second_per_gpu:3336.2483024753146 
+Step 536 | loss:1.1073501110076904 lr:1.0421729117448334e-06 tokens_per_second_per_gpu:3675.8114899776433 
+Step 537 | loss:1.002111554145813 lr:1.0234121782545621e-06 tokens_per_second_per_gpu:3456.2249516746747 
+Step 538 | loss:0.988724946975708 lr:1.0048094716167097e-06 tokens_per_second_per_gpu:3571.358149202487 
+Step 539 | loss:1.0480667352676392 lr:9.863652457318939e-07 tokens_per_second_per_gpu:3861.236977577604 
+Step 540 | loss:1.0037941932678223 lr:9.68079950633841e-07 tokens_per_second_per_gpu:3795.781462671837 
+Step 541 | loss:1.046254277229309 lr:9.499540324784137e-07 tokens_per_second_per_gpu:3805.412799974102 
+Step 542 | loss:1.0471158027648926 lr:9.319879335327244e-07 tokens_per_second_per_gpu:3803.6177569269944 
+Step 543 | loss:1.0406914949417114 lr:9.141820921643429e-07 tokens_per_second_per_gpu:3697.1148852757697 
+Step 544 | loss:1.0133854150772095 lr:8.965369428306025e-07 tokens_per_second_per_gpu:2798.1724809223538 
+Step 545 | loss:0.9686719179153442 lr:8.790529160679947e-07 tokens_per_second_per_gpu:3461.1321675655404 
+Step 546 | loss:0.9938296675682068 lr:8.617304384816716e-07 tokens_per_second_per_gpu:3745.8216708473556 
+Step 547 | loss:1.035244345664978 lr:8.445699327350281e-07 tokens_per_second_per_gpu:3470.7464874423918 
+Step 548 | loss:1.0502572059631348 lr:8.275718175393959e-07 tokens_per_second_per_gpu:3148.8280613056 
+Step 549 | loss:0.9388608336448669 lr:8.10736507643825e-07 tokens_per_second_per_gpu:3426.3978598698536 
+Step 550 | loss:1.0639617443084717 lr:7.940644138249606e-07 tokens_per_second_per_gpu:3617.338930211458 
+Step 551 | loss:1.0359141826629639 lr:7.77555942877031e-07 tokens_per_second_per_gpu:3569.7245310019453 
+Step 552 | loss:1.0284898281097412 lr:7.612114976019013e-07 tokens_per_second_per_gpu:4029.1180368589817 
+Step 553 | loss:1.0912672281265259 lr:7.450314767992725e-07 tokens_per_second_per_gpu:3319.4160159719727 
+Step 554 | loss:1.0639758110046387 lr:7.290162752569235e-07 tokens_per_second_per_gpu:3216.1711606947674 
+Step 555 | loss:0.9887262582778931 lr:7.131662837411021e-07 tokens_per_second_per_gpu:3598.567838296343 
+Step 556 | loss:1.119757890701294 lr:6.974818889869706e-07 tokens_per_second_per_gpu:3828.479312234801 
+Step 557 | loss:0.9670307636260986 lr:6.819634736891881e-07 tokens_per_second_per_gpu:3205.8416970935805 
+Step 558 | loss:0.8852719664573669 lr:6.666114164925519e-07 tokens_per_second_per_gpu:3146.806660385405 
+Step 559 | loss:0.9563117623329163 lr:6.51426091982782e-07 tokens_per_second_per_gpu:3621.68107570963 
+Step 560 | loss:1.1427892446517944 lr:6.364078706773616e-07 tokens_per_second_per_gpu:2813.4747436797943 
+Step 561 | loss:1.0382424592971802 lr:6.215571190165073e-07 tokens_per_second_per_gpu:3323.0282890047843 
+Step 562 | loss:1.0987064838409424 lr:6.068741993542251e-07 tokens_per_second_per_gpu:3476.205922040534 
+Step 563 | loss:1.0931929349899292 lr:5.923594699494683e-07 tokens_per_second_per_gpu:3470.4852308972336 
+Step 564 | loss:0.9598541259765625 lr:5.780132849573988e-07 tokens_per_second_per_gpu:3137.686342427438 
+Step 565 | loss:0.9403815865516663 lr:5.638359944207421e-07 tokens_per_second_per_gpu:3236.405721258059 
+Step 566 | loss:0.9363535046577454 lr:5.498279442612497e-07 tokens_per_second_per_gpu:3329.627606082326 
+Step 567 | loss:1.1313518285751343 lr:5.359894762712558e-07 tokens_per_second_per_gpu:3055.2431705401013 
+Step 568 | loss:1.015571117401123 lr:5.223209281053415e-07 tokens_per_second_per_gpu:3085.9648675924795 
+Step 569 | loss:1.0932440757751465 lr:5.088226332720916e-07 tokens_per_second_per_gpu:3736.1741995184375 
+Step 570 | loss:1.132112979888916 lr:4.954949211259599e-07 tokens_per_second_per_gpu:3456.21209578924 
+Step 571 | loss:1.0685855150222778 lr:4.823381168592328e-07 tokens_per_second_per_gpu:3930.7959672235484 
+Step 572 | loss:0.9238698482513428 lr:4.693525414940933e-07 tokens_per_second_per_gpu:3142.089445727256 
+Step 573 | loss:0.9899613857269287 lr:4.565385118747922e-07 tokens_per_second_per_gpu:3651.4091222282746 
+Step 574 | loss:1.131011962890625 lr:4.4389634065990866e-07 tokens_per_second_per_gpu:3602.03410991543 
+Step 575 | loss:1.0533106327056885 lr:4.31426336314735e-07 tokens_per_second_per_gpu:3126.586366047976 
+Step 576 | loss:0.990755558013916 lr:4.191288031037316e-07 tokens_per_second_per_gpu:2938.069229206343 
+Step 577 | loss:1.085005521774292 lr:4.0700404108312493e-07 tokens_per_second_per_gpu:3634.120014496565 
+Step 578 | loss:1.0190842151641846 lr:3.9505234609356455e-07 tokens_per_second_per_gpu:3508.693290240972 
+Step 579 | loss:1.1383253335952759 lr:3.832740097529236e-07 tokens_per_second_per_gpu:3784.3207549193335 
+Step 580 | loss:1.0955651998519897 lr:3.7166931944916713e-07 tokens_per_second_per_gpu:3443.295664825699 
+Step 581 | loss:1.020769715309143 lr:3.602385583333537e-07 tokens_per_second_per_gpu:3670.778095752044 
+Step 582 | loss:0.9577903747558594 lr:3.4898200531271796e-07 tokens_per_second_per_gpu:3453.753723845044 
+Step 583 | loss:1.0107300281524658 lr:3.3789993504386747e-07 tokens_per_second_per_gpu:3322.5018350298633 
+Step 584 | loss:1.1570929288864136 lr:3.2699261792608183e-07 tokens_per_second_per_gpu:3546.864382690297 
+Step 585 | loss:0.9978330135345459 lr:3.162603200947156e-07 tokens_per_second_per_gpu:3258.2812026952033 
+Step 586 | loss:1.0092127323150635 lr:3.057033034147028e-07 tokens_per_second_per_gpu:3262.423307313303 
+Step 587 | loss:0.9077081680297852 lr:2.953218254741699e-07 tokens_per_second_per_gpu:3697.3960382773657 
+Step 588 | loss:1.018717885017395 lr:2.8511613957814655e-07 tokens_per_second_per_gpu:3288.5442807398276 
+Step 589 | loss:1.0892322063446045 lr:2.7508649474239145e-07 tokens_per_second_per_gpu:3393.0489554763767 
+Step 590 | loss:1.0539475679397583 lr:2.6523313568731026e-07 tokens_per_second_per_gpu:3643.0417749516805 
+Step 591 | loss:1.033966064453125 lr:2.555563028319885e-07 tokens_per_second_per_gpu:3534.4036121090435 
+Step 592 | loss:1.0088396072387695 lr:2.46056232288322e-07 tokens_per_second_per_gpu:3681.445022735837 
+Step 593 | loss:1.0584028959274292 lr:2.3673315585526072e-07 tokens_per_second_per_gpu:3240.2921371564175 
+Step 594 | loss:1.0147311687469482 lr:2.2758730101314684e-07 tokens_per_second_per_gpu:3359.430877297497 
+Step 595 | loss:1.1457427740097046 lr:2.1861889091817133e-07 tokens_per_second_per_gpu:3525.4677911202753 
+Step 596 | loss:1.1605221033096313 lr:2.0982814439691939e-07 tokens_per_second_per_gpu:3453.302876809475 
+Step 597 | loss:1.0375539064407349 lr:2.0121527594104295e-07 tokens_per_second_per_gpu:3577.2309087051176 
+Step 598 | loss:1.0658079385757446 lr:1.9278049570201654e-07 tokens_per_second_per_gpu:3381.451046136403 
+Step 599 | loss:1.0220245122909546 lr:1.8452400948601816e-07 tokens_per_second_per_gpu:3823.4144363605037 
+Step 600 | loss:0.9951660633087158 lr:1.7644601874889894e-07 tokens_per_second_per_gpu:3713.609461806826 
+Step 601 | loss:1.119286060333252 lr:1.6854672059127635e-07 tokens_per_second_per_gpu:3864.777481244428 
+Step 602 | loss:0.9337589740753174 lr:1.6082630775371976e-07 tokens_per_second_per_gpu:3744.03231292111 
+Step 603 | loss:1.072335958480835 lr:1.532849686120491e-07 tokens_per_second_per_gpu:3751.680025273069 
+Step 604 | loss:1.048313021659851 lr:1.459228871727386e-07 tokens_per_second_per_gpu:3755.7832090988786 
+Step 605 | loss:1.1349972486495972 lr:1.3874024306842453e-07 tokens_per_second_per_gpu:3677.1697050059897 
+Step 606 | loss:1.0434086322784424 lr:1.3173721155352868e-07 tokens_per_second_per_gpu:3393.0209085636616 
+Step 607 | loss:1.1098034381866455 lr:1.249139634999752e-07 tokens_per_second_per_gpu:3051.599689491925 
+Step 608 | loss:1.0198705196380615 lr:1.1827066539302378e-07 tokens_per_second_per_gpu:3692.11290536148 
+Step 609 | loss:1.0850074291229248 lr:1.1180747932721142e-07 tokens_per_second_per_gpu:3417.4471054705496 
+Step 610 | loss:1.187023639678955 lr:1.055245630023896e-07 tokens_per_second_per_gpu:3640.1749871015163 
+Step 611 | loss:1.0894136428833008 lr:9.942206971988416e-08 tokens_per_second_per_gpu:3595.7152958928796 
+Step 612 | loss:1.0794929265975952 lr:9.350014837874899e-08 tokens_per_second_per_gpu:3323.2475499204224 
+Step 613 | loss:0.9975882768630981 lr:8.775894347213654e-08 tokens_per_second_per_gpu:3418.980031398386 
+Step 614 | loss:1.220999836921692 lr:8.219859508376975e-08 tokens_per_second_per_gpu:3446.3651662681623 
+Step 615 | loss:1.0748283863067627 lr:7.681923888452902e-08 tokens_per_second_per_gpu:3373.2525132787046 
+Step 616 | loss:1.1435097455978394 lr:7.162100612913308e-08 tokens_per_second_per_gpu:3363.609081700875 
+Step 617 | loss:1.0882154703140259 lr:6.660402365294499e-08 tokens_per_second_per_gpu:3023.7098014074804 
+Step 618 | loss:0.9880639314651489 lr:6.176841386887205e-08 tokens_per_second_per_gpu:3988.401290174963 
+Step 619 | loss:1.1591405868530273 lr:5.71142947643824e-08 tokens_per_second_per_gpu:3171.9719503194547 
+Step 620 | loss:1.0383414030075073 lr:5.264177989862312e-08 tokens_per_second_per_gpu:3086.0139727006417 
+Step 621 | loss:1.1537160873413086 lr:4.8350978399650804e-08 tokens_per_second_per_gpu:3369.979075784975 
+Step 622 | loss:1.0627126693725586 lr:4.424199496177117e-08 tokens_per_second_per_gpu:3436.2757606077344 
+Step 623 | loss:1.0991474390029907 lr:4.0314929842979466e-08 tokens_per_second_per_gpu:3929.0458952761765 
+Step 624 | loss:1.0756220817565918 lr:3.6569878862519055e-08 tokens_per_second_per_gpu:3385.603493456278 
+Step 625 | loss:1.0126550197601318 lr:3.300693339854083e-08 tokens_per_second_per_gpu:3796.7319365547914 
+Step 626 | loss:1.0999494791030884 lr:2.9626180385873282e-08 tokens_per_second_per_gpu:3491.292618600503 
+Step 627 | loss:1.0032886266708374 lr:2.64277023139034e-08 tokens_per_second_per_gpu:3434.2746828936133 
+Step 628 | loss:0.9731086492538452 lr:2.3411577224563273e-08 tokens_per_second_per_gpu:3992.9430945768536 
+Step 629 | loss:1.029701590538025 lr:2.0577878710424104e-08 tokens_per_second_per_gpu:3954.800918281961 
+Step 630 | loss:0.9682028889656067 lr:1.7926675912902644e-08 tokens_per_second_per_gpu:3079.0321478215965 
+Step 631 | loss:1.2128889560699463 lr:1.54580335205734e-08 tokens_per_second_per_gpu:2766.3162516442967 
+Step 632 | loss:1.0580332279205322 lr:1.3172011767589865e-08 tokens_per_second_per_gpu:3414.5038307314135 
+Step 633 | loss:0.9746403098106384 lr:1.1068666432215713e-08 tokens_per_second_per_gpu:3261.521524950618 
+Step 634 | loss:0.9512935280799866 lr:9.148048835462552e-09 tokens_per_second_per_gpu:3716.130203693814 
+Step 635 | loss:0.9694274663925171 lr:7.410205839840922e-09 tokens_per_second_per_gpu:3638.283430143105 
+Step 636 | loss:0.9943816065788269 lr:5.8551798482112116e-09 tokens_per_second_per_gpu:3984.6815410831828 
+Step 637 | loss:1.019755244255066 lr:4.483008802754485e-09 tokens_per_second_per_gpu:3426.6431524886534 
+Step 638 | loss:0.9445756673812866 lr:3.2937261840423894e-09 tokens_per_second_per_gpu:3651.1875501336294 
+Step 639 | loss:1.0361696481704712 lr:2.2873610102253083e-09 tokens_per_second_per_gpu:2899.609547551191 
+Step 640 | loss:1.0265421867370605 lr:1.4639378363187629e-09 tokens_per_second_per_gpu:3146.508069944556 
+Step 641 | loss:0.9879869222640991 lr:8.234767536080545e-10 tokens_per_second_per_gpu:3718.1335290345437 
+Step 642 | loss:1.0564134120941162 lr:3.659933891569933e-10 tokens_per_second_per_gpu:3539.4915099356376 
+Step 643 | loss:1.1678091287612915 lr:9.149890542570205e-11 tokens_per_second_per_gpu:2749.888827785128 
+Step 644 | loss:0.9817880392074585 lr:0.0 tokens_per_second_per_gpu:3425.3032445871813