| { | |
| "best_metric": null, | |
| "best_model_checkpoint": null, | |
| "epoch": 4.935064935064935, | |
| "eval_steps": 500, | |
| "global_step": 95, | |
| "is_hyper_param_search": false, | |
| "is_local_process_zero": true, | |
| "is_world_process_zero": true, | |
| "log_history": [ | |
| { | |
| "epoch": 0.05194805194805195, | |
| "grad_norm": 4.959436893463135, | |
| "learning_rate": 2e-05, | |
| "loss": 2.0897, | |
| "step": 1 | |
| }, | |
| { | |
| "epoch": 0.1038961038961039, | |
| "grad_norm": 3.942784547805786, | |
| "learning_rate": 4e-05, | |
| "loss": 2.3866, | |
| "step": 2 | |
| }, | |
| { | |
| "epoch": 0.15584415584415584, | |
| "grad_norm": 4.221513748168945, | |
| "learning_rate": 6e-05, | |
| "loss": 2.4587, | |
| "step": 3 | |
| }, | |
| { | |
| "epoch": 0.2077922077922078, | |
| "grad_norm": 3.284791946411133, | |
| "learning_rate": 8e-05, | |
| "loss": 2.1799, | |
| "step": 4 | |
| }, | |
| { | |
| "epoch": 0.2597402597402597, | |
| "grad_norm": 5.71047306060791, | |
| "learning_rate": 0.0001, | |
| "loss": 2.1779, | |
| "step": 5 | |
| }, | |
| { | |
| "epoch": 0.3116883116883117, | |
| "grad_norm": 3.1526741981506348, | |
| "learning_rate": 0.00012, | |
| "loss": 2.1932, | |
| "step": 6 | |
| }, | |
| { | |
| "epoch": 0.36363636363636365, | |
| "grad_norm": 3.1342685222625732, | |
| "learning_rate": 0.00014, | |
| "loss": 1.4442, | |
| "step": 7 | |
| }, | |
| { | |
| "epoch": 0.4155844155844156, | |
| "grad_norm": 5.338088035583496, | |
| "learning_rate": 0.00016, | |
| "loss": 2.0, | |
| "step": 8 | |
| }, | |
| { | |
| "epoch": 0.4675324675324675, | |
| "grad_norm": 4.116346836090088, | |
| "learning_rate": 0.00018, | |
| "loss": 1.5286, | |
| "step": 9 | |
| }, | |
| { | |
| "epoch": 0.5194805194805194, | |
| "grad_norm": 2.5058088302612305, | |
| "learning_rate": 0.0002, | |
| "loss": 1.4238, | |
| "step": 10 | |
| }, | |
| { | |
| "epoch": 0.5714285714285714, | |
| "grad_norm": 4.636209487915039, | |
| "learning_rate": 0.0001999317060143023, | |
| "loss": 1.511, | |
| "step": 11 | |
| }, | |
| { | |
| "epoch": 0.6233766233766234, | |
| "grad_norm": 3.1061031818389893, | |
| "learning_rate": 0.00019972691733857883, | |
| "loss": 1.3241, | |
| "step": 12 | |
| }, | |
| { | |
| "epoch": 0.6753246753246753, | |
| "grad_norm": 2.3636441230773926, | |
| "learning_rate": 0.0001993859136895274, | |
| "loss": 1.044, | |
| "step": 13 | |
| }, | |
| { | |
| "epoch": 0.7272727272727273, | |
| "grad_norm": 2.5445024967193604, | |
| "learning_rate": 0.0001989091608371146, | |
| "loss": 0.9548, | |
| "step": 14 | |
| }, | |
| { | |
| "epoch": 0.7792207792207793, | |
| "grad_norm": 2.371872901916504, | |
| "learning_rate": 0.0001982973099683902, | |
| "loss": 1.0614, | |
| "step": 15 | |
| }, | |
| { | |
| "epoch": 0.8311688311688312, | |
| "grad_norm": 5.0257039070129395, | |
| "learning_rate": 0.00019755119679804367, | |
| "loss": 1.1408, | |
| "step": 16 | |
| }, | |
| { | |
| "epoch": 0.8831168831168831, | |
| "grad_norm": 1.9052318334579468, | |
| "learning_rate": 0.00019667184042691875, | |
| "loss": 0.8993, | |
| "step": 17 | |
| }, | |
| { | |
| "epoch": 0.935064935064935, | |
| "grad_norm": 2.2592344284057617, | |
| "learning_rate": 0.0001956604419500441, | |
| "loss": 1.0319, | |
| "step": 18 | |
| }, | |
| { | |
| "epoch": 0.987012987012987, | |
| "grad_norm": 2.264328718185425, | |
| "learning_rate": 0.00019451838281608197, | |
| "loss": 1.483, | |
| "step": 19 | |
| }, | |
| { | |
| "epoch": 1.0389610389610389, | |
| "grad_norm": 4.711090564727783, | |
| "learning_rate": 0.00019324722294043558, | |
| "loss": 1.4984, | |
| "step": 20 | |
| }, | |
| { | |
| "epoch": 1.0909090909090908, | |
| "grad_norm": 1.5374799966812134, | |
| "learning_rate": 0.00019184869857459232, | |
| "loss": 0.7076, | |
| "step": 21 | |
| }, | |
| { | |
| "epoch": 1.1428571428571428, | |
| "grad_norm": 2.26493501663208, | |
| "learning_rate": 0.0001903247199346129, | |
| "loss": 1.1543, | |
| "step": 22 | |
| }, | |
| { | |
| "epoch": 1.1948051948051948, | |
| "grad_norm": 1.51936674118042, | |
| "learning_rate": 0.0001886773685920062, | |
| "loss": 0.7064, | |
| "step": 23 | |
| }, | |
| { | |
| "epoch": 1.2467532467532467, | |
| "grad_norm": 2.1054816246032715, | |
| "learning_rate": 0.00018690889463055283, | |
| "loss": 0.8401, | |
| "step": 24 | |
| }, | |
| { | |
| "epoch": 1.2987012987012987, | |
| "grad_norm": 2.4671549797058105, | |
| "learning_rate": 0.00018502171357296144, | |
| "loss": 1.0385, | |
| "step": 25 | |
| }, | |
| { | |
| "epoch": 1.3506493506493507, | |
| "grad_norm": 1.6618045568466187, | |
| "learning_rate": 0.00018301840308155507, | |
| "loss": 0.5543, | |
| "step": 26 | |
| }, | |
| { | |
| "epoch": 1.4025974025974026, | |
| "grad_norm": 1.7877517938613892, | |
| "learning_rate": 0.00018090169943749476, | |
| "loss": 0.5464, | |
| "step": 27 | |
| }, | |
| { | |
| "epoch": 1.4545454545454546, | |
| "grad_norm": 3.1368205547332764, | |
| "learning_rate": 0.00017867449380334834, | |
| "loss": 1.2296, | |
| "step": 28 | |
| }, | |
| { | |
| "epoch": 1.5064935064935066, | |
| "grad_norm": 1.902140498161316, | |
| "learning_rate": 0.00017633982827411032, | |
| "loss": 0.8562, | |
| "step": 29 | |
| }, | |
| { | |
| "epoch": 1.5584415584415585, | |
| "grad_norm": 1.6555312871932983, | |
| "learning_rate": 0.00017390089172206592, | |
| "loss": 0.7186, | |
| "step": 30 | |
| }, | |
| { | |
| "epoch": 1.6103896103896105, | |
| "grad_norm": 1.526365041732788, | |
| "learning_rate": 0.00017136101544117525, | |
| "loss": 0.6501, | |
| "step": 31 | |
| }, | |
| { | |
| "epoch": 1.6623376623376624, | |
| "grad_norm": 1.7592672109603882, | |
| "learning_rate": 0.00016872366859692627, | |
| "loss": 0.9721, | |
| "step": 32 | |
| }, | |
| { | |
| "epoch": 1.7142857142857144, | |
| "grad_norm": 1.7448221445083618, | |
| "learning_rate": 0.0001659924534878723, | |
| "loss": 0.6916, | |
| "step": 33 | |
| }, | |
| { | |
| "epoch": 1.7662337662337664, | |
| "grad_norm": 2.216275930404663, | |
| "learning_rate": 0.0001631711006253251, | |
| "loss": 0.8664, | |
| "step": 34 | |
| }, | |
| { | |
| "epoch": 1.8181818181818183, | |
| "grad_norm": 1.5474787950515747, | |
| "learning_rate": 0.00016026346363792567, | |
| "loss": 0.5205, | |
| "step": 35 | |
| }, | |
| { | |
| "epoch": 1.87012987012987, | |
| "grad_norm": 2.3516626358032227, | |
| "learning_rate": 0.00015727351400805052, | |
| "loss": 1.0247, | |
| "step": 36 | |
| }, | |
| { | |
| "epoch": 1.922077922077922, | |
| "grad_norm": 1.921420931816101, | |
| "learning_rate": 0.00015420533564724495, | |
| "loss": 0.9075, | |
| "step": 37 | |
| }, | |
| { | |
| "epoch": 1.974025974025974, | |
| "grad_norm": 1.9142687320709229, | |
| "learning_rate": 0.0001510631193180907, | |
| "loss": 0.8456, | |
| "step": 38 | |
| }, | |
| { | |
| "epoch": 2.0259740259740258, | |
| "grad_norm": 2.5155398845672607, | |
| "learning_rate": 0.00014785115691012864, | |
| "loss": 0.7963, | |
| "step": 39 | |
| }, | |
| { | |
| "epoch": 2.0779220779220777, | |
| "grad_norm": 1.6434123516082764, | |
| "learning_rate": 0.00014457383557765386, | |
| "loss": 0.7432, | |
| "step": 40 | |
| }, | |
| { | |
| "epoch": 2.1298701298701297, | |
| "grad_norm": 1.6011264324188232, | |
| "learning_rate": 0.00014123563174739037, | |
| "loss": 0.5455, | |
| "step": 41 | |
| }, | |
| { | |
| "epoch": 2.1818181818181817, | |
| "grad_norm": 1.7301688194274902, | |
| "learning_rate": 0.00013784110500423104, | |
| "loss": 0.4415, | |
| "step": 42 | |
| }, | |
| { | |
| "epoch": 2.2337662337662336, | |
| "grad_norm": 1.334765911102295, | |
| "learning_rate": 0.00013439489186339282, | |
| "loss": 0.3515, | |
| "step": 43 | |
| }, | |
| { | |
| "epoch": 2.2857142857142856, | |
| "grad_norm": 1.9331822395324707, | |
| "learning_rate": 0.00013090169943749476, | |
| "loss": 0.6585, | |
| "step": 44 | |
| }, | |
| { | |
| "epoch": 2.3376623376623376, | |
| "grad_norm": 1.5188448429107666, | |
| "learning_rate": 0.0001273662990072083, | |
| "loss": 0.3202, | |
| "step": 45 | |
| }, | |
| { | |
| "epoch": 2.3896103896103895, | |
| "grad_norm": 2.288816213607788, | |
| "learning_rate": 0.00012379351950426187, | |
| "loss": 0.6965, | |
| "step": 46 | |
| }, | |
| { | |
| "epoch": 2.4415584415584415, | |
| "grad_norm": 1.356819987297058, | |
| "learning_rate": 0.00012018824091570103, | |
| "loss": 0.2039, | |
| "step": 47 | |
| }, | |
| { | |
| "epoch": 2.4935064935064934, | |
| "grad_norm": 2.0376572608947754, | |
| "learning_rate": 0.000116555387618413, | |
| "loss": 0.4097, | |
| "step": 48 | |
| }, | |
| { | |
| "epoch": 2.5454545454545454, | |
| "grad_norm": 1.3941322565078735, | |
| "learning_rate": 0.00011289992165302035, | |
| "loss": 0.318, | |
| "step": 49 | |
| }, | |
| { | |
| "epoch": 2.5974025974025974, | |
| "grad_norm": 1.9851269721984863, | |
| "learning_rate": 0.00010922683594633021, | |
| "loss": 0.4971, | |
| "step": 50 | |
| }, | |
| { | |
| "epoch": 2.6493506493506493, | |
| "grad_norm": 1.670423984527588, | |
| "learning_rate": 0.000105541147491597, | |
| "loss": 0.3796, | |
| "step": 51 | |
| }, | |
| { | |
| "epoch": 2.7012987012987013, | |
| "grad_norm": 1.6468473672866821, | |
| "learning_rate": 0.00010184789049591299, | |
| "loss": 0.5085, | |
| "step": 52 | |
| }, | |
| { | |
| "epoch": 2.7532467532467533, | |
| "grad_norm": 1.2429389953613281, | |
| "learning_rate": 9.815210950408704e-05, | |
| "loss": 0.2477, | |
| "step": 53 | |
| }, | |
| { | |
| "epoch": 2.8051948051948052, | |
| "grad_norm": 2.051650047302246, | |
| "learning_rate": 9.4458852508403e-05, | |
| "loss": 0.6422, | |
| "step": 54 | |
| }, | |
| { | |
| "epoch": 2.857142857142857, | |
| "grad_norm": 1.2772984504699707, | |
| "learning_rate": 9.077316405366981e-05, | |
| "loss": 0.2495, | |
| "step": 55 | |
| }, | |
| { | |
| "epoch": 2.909090909090909, | |
| "grad_norm": 1.7530468702316284, | |
| "learning_rate": 8.710007834697969e-05, | |
| "loss": 0.4244, | |
| "step": 56 | |
| }, | |
| { | |
| "epoch": 2.961038961038961, | |
| "grad_norm": 1.568015694618225, | |
| "learning_rate": 8.344461238158699e-05, | |
| "loss": 0.3102, | |
| "step": 57 | |
| }, | |
| { | |
| "epoch": 3.012987012987013, | |
| "grad_norm": 3.1359732151031494, | |
| "learning_rate": 7.9811759084299e-05, | |
| "loss": 0.978, | |
| "step": 58 | |
| }, | |
| { | |
| "epoch": 3.064935064935065, | |
| "grad_norm": 1.2793606519699097, | |
| "learning_rate": 7.620648049573815e-05, | |
| "loss": 0.2023, | |
| "step": 59 | |
| }, | |
| { | |
| "epoch": 3.116883116883117, | |
| "grad_norm": 0.8895975947380066, | |
| "learning_rate": 7.263370099279172e-05, | |
| "loss": 0.1311, | |
| "step": 60 | |
| }, | |
| { | |
| "epoch": 3.168831168831169, | |
| "grad_norm": 1.6454602479934692, | |
| "learning_rate": 6.909830056250527e-05, | |
| "loss": 0.2349, | |
| "step": 61 | |
| }, | |
| { | |
| "epoch": 3.220779220779221, | |
| "grad_norm": 3.0135691165924072, | |
| "learning_rate": 6.560510813660719e-05, | |
| "loss": 0.282, | |
| "step": 62 | |
| }, | |
| { | |
| "epoch": 3.2727272727272725, | |
| "grad_norm": 2.029156446456909, | |
| "learning_rate": 6.215889499576898e-05, | |
| "loss": 0.2724, | |
| "step": 63 | |
| }, | |
| { | |
| "epoch": 3.324675324675325, | |
| "grad_norm": 1.4129441976547241, | |
| "learning_rate": 5.876436825260967e-05, | |
| "loss": 0.2536, | |
| "step": 64 | |
| }, | |
| { | |
| "epoch": 3.3766233766233764, | |
| "grad_norm": 0.968977153301239, | |
| "learning_rate": 5.542616442234618e-05, | |
| "loss": 0.1043, | |
| "step": 65 | |
| }, | |
| { | |
| "epoch": 3.4285714285714284, | |
| "grad_norm": 2.872439384460449, | |
| "learning_rate": 5.214884308987136e-05, | |
| "loss": 0.3041, | |
| "step": 66 | |
| }, | |
| { | |
| "epoch": 3.4805194805194803, | |
| "grad_norm": 2.175682306289673, | |
| "learning_rate": 4.893688068190932e-05, | |
| "loss": 0.2227, | |
| "step": 67 | |
| }, | |
| { | |
| "epoch": 3.5324675324675323, | |
| "grad_norm": 1.2863298654556274, | |
| "learning_rate": 4.5794664352755055e-05, | |
| "loss": 0.1234, | |
| "step": 68 | |
| }, | |
| { | |
| "epoch": 3.5844155844155843, | |
| "grad_norm": 1.820629596710205, | |
| "learning_rate": 4.272648599194948e-05, | |
| "loss": 0.1685, | |
| "step": 69 | |
| }, | |
| { | |
| "epoch": 3.6363636363636362, | |
| "grad_norm": 1.7225141525268555, | |
| "learning_rate": 3.973653636207437e-05, | |
| "loss": 0.3388, | |
| "step": 70 | |
| }, | |
| { | |
| "epoch": 3.688311688311688, | |
| "grad_norm": 3.239525556564331, | |
| "learning_rate": 3.682889937467493e-05, | |
| "loss": 0.2992, | |
| "step": 71 | |
| }, | |
| { | |
| "epoch": 3.74025974025974, | |
| "grad_norm": 1.7865426540374756, | |
| "learning_rate": 3.400754651212776e-05, | |
| "loss": 0.1705, | |
| "step": 72 | |
| }, | |
| { | |
| "epoch": 3.792207792207792, | |
| "grad_norm": 1.409381628036499, | |
| "learning_rate": 3.1276331403073735e-05, | |
| "loss": 0.1688, | |
| "step": 73 | |
| }, | |
| { | |
| "epoch": 3.844155844155844, | |
| "grad_norm": 2.8873255252838135, | |
| "learning_rate": 2.8638984558824777e-05, | |
| "loss": 0.2449, | |
| "step": 74 | |
| }, | |
| { | |
| "epoch": 3.896103896103896, | |
| "grad_norm": 1.371756672859192, | |
| "learning_rate": 2.6099108277934103e-05, | |
| "loss": 0.1109, | |
| "step": 75 | |
| }, | |
| { | |
| "epoch": 3.948051948051948, | |
| "grad_norm": 1.8516228199005127, | |
| "learning_rate": 2.36601717258897e-05, | |
| "loss": 0.2749, | |
| "step": 76 | |
| }, | |
| { | |
| "epoch": 4.0, | |
| "grad_norm": 5.486507892608643, | |
| "learning_rate": 2.132550619665168e-05, | |
| "loss": 0.3419, | |
| "step": 77 | |
| }, | |
| { | |
| "epoch": 4.0519480519480515, | |
| "grad_norm": 1.2953009605407715, | |
| "learning_rate": 1.9098300562505266e-05, | |
| "loss": 0.169, | |
| "step": 78 | |
| }, | |
| { | |
| "epoch": 4.103896103896104, | |
| "grad_norm": 4.110390663146973, | |
| "learning_rate": 1.6981596918444953e-05, | |
| "loss": 0.1386, | |
| "step": 79 | |
| }, | |
| { | |
| "epoch": 4.1558441558441555, | |
| "grad_norm": 1.1933214664459229, | |
| "learning_rate": 1.4978286427038601e-05, | |
| "loss": 0.1422, | |
| "step": 80 | |
| }, | |
| { | |
| "epoch": 4.207792207792208, | |
| "grad_norm": 1.1193217039108276, | |
| "learning_rate": 1.3091105369447165e-05, | |
| "loss": 0.1003, | |
| "step": 81 | |
| }, | |
| { | |
| "epoch": 4.259740259740259, | |
| "grad_norm": 1.461154818534851, | |
| "learning_rate": 1.1322631407993811e-05, | |
| "loss": 0.2098, | |
| "step": 82 | |
| }, | |
| { | |
| "epoch": 4.311688311688312, | |
| "grad_norm": 1.960875153541565, | |
| "learning_rate": 9.675280065387116e-06, | |
| "loss": 0.1864, | |
| "step": 83 | |
| }, | |
| { | |
| "epoch": 4.363636363636363, | |
| "grad_norm": 1.9848136901855469, | |
| "learning_rate": 8.151301425407699e-06, | |
| "loss": 0.1859, | |
| "step": 84 | |
| }, | |
| { | |
| "epoch": 4.415584415584416, | |
| "grad_norm": 0.6564483046531677, | |
| "learning_rate": 6.75277705956443e-06, | |
| "loss": 0.0753, | |
| "step": 85 | |
| }, | |
| { | |
| "epoch": 4.467532467532467, | |
| "grad_norm": 0.8362023234367371, | |
| "learning_rate": 5.481617183918053e-06, | |
| "loss": 0.1173, | |
| "step": 86 | |
| }, | |
| { | |
| "epoch": 4.51948051948052, | |
| "grad_norm": 2.1371753215789795, | |
| "learning_rate": 4.339558049955927e-06, | |
| "loss": 0.2897, | |
| "step": 87 | |
| }, | |
| { | |
| "epoch": 4.571428571428571, | |
| "grad_norm": 1.301545262336731, | |
| "learning_rate": 3.3281595730812575e-06, | |
| "loss": 0.1948, | |
| "step": 88 | |
| }, | |
| { | |
| "epoch": 4.623376623376624, | |
| "grad_norm": 1.0829373598098755, | |
| "learning_rate": 2.4488032019563402e-06, | |
| "loss": 0.1707, | |
| "step": 89 | |
| }, | |
| { | |
| "epoch": 4.675324675324675, | |
| "grad_norm": 0.8257635235786438, | |
| "learning_rate": 1.7026900316098215e-06, | |
| "loss": 0.0545, | |
| "step": 90 | |
| }, | |
| { | |
| "epoch": 4.7272727272727275, | |
| "grad_norm": 1.0322595834732056, | |
| "learning_rate": 1.0908391628854041e-06, | |
| "loss": 0.1287, | |
| "step": 91 | |
| }, | |
| { | |
| "epoch": 4.779220779220779, | |
| "grad_norm": 1.001582384109497, | |
| "learning_rate": 6.140863104726391e-07, | |
| "loss": 0.1282, | |
| "step": 92 | |
| }, | |
| { | |
| "epoch": 4.8311688311688314, | |
| "grad_norm": 1.6879202127456665, | |
| "learning_rate": 2.7308266142119785e-07, | |
| "loss": 0.1124, | |
| "step": 93 | |
| }, | |
| { | |
| "epoch": 4.883116883116883, | |
| "grad_norm": 1.0578956604003906, | |
| "learning_rate": 6.829398569770939e-08, | |
| "loss": 0.1178, | |
| "step": 94 | |
| }, | |
| { | |
| "epoch": 4.935064935064935, | |
| "grad_norm": 1.2711213827133179, | |
| "learning_rate": 0.0, | |
| "loss": 0.0993, | |
| "step": 95 | |
| } | |
| ], | |
| "logging_steps": 1, | |
| "max_steps": 95, | |
| "num_input_tokens_seen": 0, | |
| "num_train_epochs": 5, | |
| "save_steps": 19, | |
| "stateful_callbacks": { | |
| "TrainerControl": { | |
| "args": { | |
| "should_epoch_stop": false, | |
| "should_evaluate": false, | |
| "should_log": false, | |
| "should_save": true, | |
| "should_training_stop": true | |
| }, | |
| "attributes": {} | |
| } | |
| }, | |
| "total_flos": 3.1538580107034624e+17, | |
| "train_batch_size": 1, | |
| "trial_name": null, | |
| "trial_params": null | |
| } | |