{
  "best_global_step": null,
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 4.0,
  "eval_steps": 21,
  "global_step": 108,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.037037037037037035,
      "grad_norm": 0.28136003017425537,
      "learning_rate": 0.0,
      "loss": 1.6494140625,
      "step": 1
    },
    {
      "epoch": 0.07407407407407407,
      "grad_norm": 0.2723458707332611,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 1.62890625,
      "step": 2
    },
    {
      "epoch": 0.1111111111111111,
      "grad_norm": 0.2669340670108795,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 1.6201171875,
      "step": 3
    },
    {
      "epoch": 0.14814814814814814,
      "grad_norm": 0.198916956782341,
      "learning_rate": 5e-05,
      "loss": 1.5537109375,
      "step": 4
    },
    {
      "epoch": 0.18518518518518517,
      "grad_norm": 0.1594628393650055,
      "learning_rate": 6.666666666666667e-05,
      "loss": 1.4970703125,
      "step": 5
    },
    {
      "epoch": 0.2222222222222222,
      "grad_norm": 0.12411235272884369,
      "learning_rate": 8.333333333333334e-05,
      "loss": 1.4560546875,
      "step": 6
    },
    {
      "epoch": 0.25925925925925924,
      "grad_norm": 0.18533363938331604,
      "learning_rate": 0.0001,
      "loss": 1.435546875,
      "step": 7
    },
    {
      "epoch": 0.2962962962962963,
      "grad_norm": 0.24133899807929993,
      "learning_rate": 9.997865738710147e-05,
      "loss": 1.404296875,
      "step": 8
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.20911315083503723,
      "learning_rate": 9.991464979316699e-05,
      "loss": 1.3662109375,
      "step": 9
    },
    {
      "epoch": 0.37037037037037035,
      "grad_norm": 0.1495620608329773,
      "learning_rate": 9.980803793327656e-05,
      "loss": 1.3505859375,
      "step": 10
    },
    {
      "epoch": 0.4074074074074074,
      "grad_norm": 0.09664365649223328,
      "learning_rate": 9.965892293523712e-05,
      "loss": 1.328125,
      "step": 11
    },
    {
      "epoch": 0.4444444444444444,
      "grad_norm": 0.08426272124052048,
      "learning_rate": 9.946744624365668e-05,
      "loss": 1.296875,
      "step": 12
    },
    {
      "epoch": 0.48148148148148145,
      "grad_norm": 0.08858562260866165,
      "learning_rate": 9.923378948577559e-05,
      "loss": 1.35791015625,
      "step": 13
    },
    {
      "epoch": 0.5185185185185185,
      "grad_norm": 0.08885785192251205,
      "learning_rate": 9.895817429918203e-05,
      "loss": 1.2822265625,
      "step": 14
    },
    {
      "epoch": 0.5555555555555556,
      "grad_norm": 0.08178187161684036,
      "learning_rate": 9.864086212157544e-05,
      "loss": 1.265625,
      "step": 15
    },
    {
      "epoch": 0.5925925925925926,
      "grad_norm": 0.06893579661846161,
      "learning_rate": 9.828215394277687e-05,
      "loss": 1.2490234375,
      "step": 16
    },
    {
      "epoch": 0.6296296296296297,
      "grad_norm": 0.05814003571867943,
      "learning_rate": 9.788239001922206e-05,
      "loss": 1.216796875,
      "step": 17
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.06850015372037888,
      "learning_rate": 9.744194955120748e-05,
      "loss": 1.2255859375,
      "step": 18
    },
    {
      "epoch": 0.7037037037037037,
      "grad_norm": 0.05855270102620125,
      "learning_rate": 9.6961250323196e-05,
      "loss": 1.205078125,
      "step": 19
    },
    {
      "epoch": 0.7407407407407407,
      "grad_norm": 0.051300644874572754,
      "learning_rate": 9.644074830752293e-05,
      "loss": 1.212890625,
      "step": 20
    },
    {
      "epoch": 0.7777777777777778,
      "grad_norm": 0.04694854095578194,
      "learning_rate": 9.588093723187857e-05,
      "loss": 1.2021484375,
      "step": 21
    },
    {
      "epoch": 0.8148148148148148,
      "grad_norm": 0.04827137663960457,
      "learning_rate": 9.528234811097782e-05,
      "loss": 1.2099609375,
      "step": 22
    },
    {
      "epoch": 0.8518518518518519,
      "grad_norm": 0.04695910960435867,
      "learning_rate": 9.46455487428603e-05,
      "loss": 1.185546875,
      "step": 23
    },
    {
      "epoch": 0.8888888888888888,
      "grad_norm": 0.04019802063703537,
      "learning_rate": 9.397114317029975e-05,
      "loss": 1.19140625,
      "step": 24
    },
    {
      "epoch": 0.8888888888888888,
      "eval_loss": 1.1943359375,
      "eval_runtime": 5.5855,
      "eval_samples_per_second": 1.253,
      "eval_steps_per_second": 0.179,
      "step": 24
    },
    {
      "epoch": 0.9259259259259259,
      "grad_norm": 0.040321558713912964,
      "learning_rate": 9.325977110783264e-05,
      "loss": 1.1689453125,
      "step": 25
    },
    {
      "epoch": 0.9629629629629629,
      "grad_norm": 0.04316447675228119,
      "learning_rate": 9.251210733495039e-05,
      "loss": 1.1708984375,
      "step": 26
    },
    {
      "epoch": 1.0,
      "grad_norm": 0.039865776896476746,
      "learning_rate": 9.172886105602998e-05,
      "loss": 1.171875,
      "step": 27
    },
    {
      "epoch": 1.037037037037037,
      "grad_norm": 0.036648839712142944,
      "learning_rate": 9.091077522761079e-05,
      "loss": 1.1787109375,
      "step": 28
    },
    {
      "epoch": 1.074074074074074,
      "grad_norm": 0.040327638387680054,
      "learning_rate": 9.005862585365517e-05,
      "loss": 1.14453125,
      "step": 29
    },
    {
      "epoch": 1.1111111111111112,
      "grad_norm": 0.03935032710433006,
      "learning_rate": 8.917322124946182e-05,
      "loss": 1.1484375,
      "step": 30
    },
    {
      "epoch": 1.1481481481481481,
      "grad_norm": 0.03480669856071472,
      "learning_rate": 8.825540127492967e-05,
      "loss": 1.130859375,
      "step": 31
    },
    {
      "epoch": 1.1851851851851851,
      "grad_norm": 0.03485634922981262,
      "learning_rate": 8.73060365378999e-05,
      "loss": 1.1259765625,
      "step": 32
    },
    {
      "epoch": 1.2222222222222223,
      "grad_norm": 0.038586605340242386,
      "learning_rate": 8.632602756833172e-05,
      "loss": 1.1298828125,
      "step": 33
    },
    {
      "epoch": 1.2592592592592593,
      "grad_norm": 0.0379546582698822,
      "learning_rate": 8.531630396409507e-05,
      "loss": 1.1083984375,
      "step": 34
    },
    {
      "epoch": 1.2962962962962963,
      "grad_norm": 0.0341765396296978,
      "learning_rate": 8.42778235091909e-05,
      "loss": 1.1015625,
      "step": 35
    },
    {
      "epoch": 1.3333333333333333,
      "grad_norm": 0.036559395492076874,
      "learning_rate": 8.32115712652348e-05,
      "loss": 1.09765625,
      "step": 36
    },
    {
      "epoch": 1.3703703703703702,
      "grad_norm": 0.03715907409787178,
      "learning_rate": 8.211855863706654e-05,
      "loss": 1.1044921875,
      "step": 37
    },
    {
      "epoch": 1.4074074074074074,
      "grad_norm": 0.03957440331578255,
      "learning_rate": 8.09998224133713e-05,
      "loss": 1.099609375,
      "step": 38
    },
    {
      "epoch": 1.4444444444444444,
      "grad_norm": 0.036663420498371124,
      "learning_rate": 7.985642378322276e-05,
      "loss": 1.0830078125,
      "step": 39
    },
    {
      "epoch": 1.4814814814814814,
      "grad_norm": 0.05821947380900383,
      "learning_rate": 7.868944732948101e-05,
      "loss": 1.1572265625,
      "step": 40
    },
    {
      "epoch": 1.5185185185185186,
      "grad_norm": 0.03782220184803009,
      "learning_rate": 7.75e-05,
      "loss": 1.0849609375,
      "step": 41
    },
    {
      "epoch": 1.5555555555555556,
      "grad_norm": 0.04202524572610855,
      "learning_rate": 7.628921005762047e-05,
      "loss": 1.083984375,
      "step": 42
    },
    {
      "epoch": 1.5925925925925926,
      "grad_norm": 0.035232119262218475,
      "learning_rate": 7.505822600994424e-05,
      "loss": 1.07421875,
      "step": 43
    },
    {
      "epoch": 1.6296296296296298,
      "grad_norm": 0.035235971212387085,
      "learning_rate": 7.380821551990525e-05,
      "loss": 1.0478515625,
      "step": 44
    },
    {
      "epoch": 1.6666666666666665,
      "grad_norm": 0.03764120116829872,
      "learning_rate": 7.254036429817058e-05,
      "loss": 1.06640625,
      "step": 45
    },
    {
      "epoch": 1.6666666666666665,
      "eval_loss": 1.0986328125,
      "eval_runtime": 5.572,
      "eval_samples_per_second": 1.256,
      "eval_steps_per_second": 0.179,
      "step": 45
    },
    {
      "epoch": 1.7037037037037037,
      "grad_norm": 0.036305468529462814,
      "learning_rate": 7.12558749784219e-05,
      "loss": 1.05078125,
      "step": 46
    },
    {
      "epoch": 1.7407407407407407,
      "grad_norm": 0.036349404603242874,
      "learning_rate": 6.995596597658468e-05,
      "loss": 1.068359375,
      "step": 47
    },
    {
      "epoch": 1.7777777777777777,
      "grad_norm": 0.034952208399772644,
      "learning_rate": 6.864187033508695e-05,
      "loss": 1.06640625,
      "step": 48
    },
    {
      "epoch": 1.8148148148148149,
      "grad_norm": 0.039700236171483994,
      "learning_rate": 6.731483455324374e-05,
      "loss": 1.076171875,
      "step": 49
    },
    {
      "epoch": 1.8518518518518519,
      "grad_norm": 0.03483971580862999,
      "learning_rate": 6.597611740487698e-05,
      "loss": 1.0595703125,
      "step": 50
    },
    {
      "epoch": 1.8888888888888888,
      "grad_norm": 0.03634239733219147,
      "learning_rate": 6.462698874429239e-05,
      "loss": 1.0673828125,
      "step": 51
    },
    {
      "epoch": 1.925925925925926,
      "grad_norm": 0.035455603152513504,
      "learning_rate": 6.326872830174567e-05,
      "loss": 1.048828125,
      "step": 52
    },
    {
      "epoch": 1.9629629629629628,
      "grad_norm": 0.035167813301086426,
      "learning_rate": 6.190262446954085e-05,
      "loss": 1.0576171875,
      "step": 53
    },
    {
      "epoch": 2.0,
      "grad_norm": 0.03683493286371231,
      "learning_rate": 6.052997307991214e-05,
      "loss": 1.05712890625,
      "step": 54
    },
    {
      "epoch": 2.037037037037037,
      "grad_norm": 0.03726612403988838,
      "learning_rate": 5.9152076175848594e-05,
      "loss": 1.0693359375,
      "step": 55
    },
    {
      "epoch": 2.074074074074074,
      "grad_norm": 0.036237966269254684,
      "learning_rate": 5.777024077602744e-05,
      "loss": 1.041015625,
      "step": 56
    },
    {
      "epoch": 2.111111111111111,
      "grad_norm": 0.037127960473299026,
      "learning_rate": 5.6385777635027684e-05,
      "loss": 1.052734375,
      "step": 57
    },
    {
      "epoch": 2.148148148148148,
      "grad_norm": 0.03780350461602211,
      "learning_rate": 5.500000000000001e-05,
      "loss": 1.0390625,
      "step": 58
    },
    {
      "epoch": 2.185185185185185,
      "grad_norm": 0.03687044233083725,
      "learning_rate": 5.361422236497235e-05,
      "loss": 1.0380859375,
      "step": 59
    },
    {
      "epoch": 2.2222222222222223,
      "grad_norm": 0.03448306396603584,
      "learning_rate": 5.2229759223972574e-05,
      "loss": 1.0419921875,
      "step": 60
    },
    {
      "epoch": 2.259259259259259,
      "grad_norm": 0.03755573183298111,
      "learning_rate": 5.0847923824151424e-05,
      "loss": 1.01806640625,
      "step": 61
    },
    {
      "epoch": 2.2962962962962963,
      "grad_norm": 0.039661504328250885,
      "learning_rate": 4.9470026920087876e-05,
      "loss": 1.017578125,
      "step": 62
    },
    {
      "epoch": 2.3333333333333335,
      "grad_norm": 0.03984743356704712,
      "learning_rate": 4.809737553045916e-05,
      "loss": 1.0224609375,
      "step": 63
    },
    {
      "epoch": 2.3703703703703702,
      "grad_norm": 0.037023186683654785,
      "learning_rate": 4.673127169825433e-05,
      "loss": 1.02392578125,
      "step": 64
    },
    {
      "epoch": 2.4074074074074074,
      "grad_norm": 0.04254316911101341,
      "learning_rate": 4.537301125570763e-05,
      "loss": 1.0244140625,
      "step": 65
    },
    {
      "epoch": 2.4444444444444446,
      "grad_norm": 0.03837042674422264,
      "learning_rate": 4.402388259512303e-05,
      "loss": 1.009765625,
      "step": 66
    },
    {
      "epoch": 2.4444444444444446,
      "eval_loss": 1.06005859375,
      "eval_runtime": 5.5677,
      "eval_samples_per_second": 1.257,
      "eval_steps_per_second": 0.18,
      "step": 66
    },
    {
      "epoch": 2.4814814814814814,
      "grad_norm": 0.04744939133524895,
      "learning_rate": 4.268516544675628e-05,
      "loss": 1.0791015625,
      "step": 67
    },
    {
      "epoch": 2.5185185185185186,
      "grad_norm": 0.043822310864925385,
      "learning_rate": 4.135812966491305e-05,
      "loss": 1.01513671875,
      "step": 68
    },
    {
      "epoch": 2.5555555555555554,
      "grad_norm": 0.04078835994005203,
      "learning_rate": 4.004403402341532e-05,
      "loss": 1.0185546875,
      "step": 69
    },
    {
      "epoch": 2.5925925925925926,
      "grad_norm": 0.03818242624402046,
      "learning_rate": 3.8744125021578126e-05,
      "loss": 1.0107421875,
      "step": 70
    },
    {
      "epoch": 2.6296296296296298,
      "grad_norm": 0.038158949464559555,
      "learning_rate": 3.7459635701829435e-05,
      "loss": 0.98486328125,
      "step": 71
    },
    {
      "epoch": 2.6666666666666665,
      "grad_norm": 0.036832571029663086,
      "learning_rate": 3.619178448009477e-05,
      "loss": 1.00732421875,
      "step": 72
    },
    {
      "epoch": 2.7037037037037037,
      "grad_norm": 0.0374872125685215,
      "learning_rate": 3.494177399005578e-05,
      "loss": 0.9931640625,
      "step": 73
    },
    {
      "epoch": 2.7407407407407405,
      "grad_norm": 0.03934762626886368,
      "learning_rate": 3.3710789942379556e-05,
      "loss": 1.0126953125,
      "step": 74
    },
    {
      "epoch": 2.7777777777777777,
      "grad_norm": 0.03733917698264122,
      "learning_rate": 3.250000000000001e-05,
      "loss": 1.01025390625,
      "step": 75
    },
    {
      "epoch": 2.814814814814815,
      "grad_norm": 0.0404835119843483,
      "learning_rate": 3.1310552670518986e-05,
      "loss": 1.0244140625,
      "step": 76
    },
    {
      "epoch": 2.851851851851852,
      "grad_norm": 0.03862153738737106,
      "learning_rate": 3.014357621677724e-05,
      "loss": 1.00830078125,
      "step": 77
    },
    {
      "epoch": 2.888888888888889,
      "grad_norm": 0.039973702281713486,
      "learning_rate": 2.90001775866287e-05,
      "loss": 1.01806640625,
      "step": 78
    },
    {
      "epoch": 2.925925925925926,
      "grad_norm": 0.03735584765672684,
      "learning_rate": 2.7881441362933468e-05,
      "loss": 1.00146484375,
      "step": 79
    },
    {
      "epoch": 2.962962962962963,
      "grad_norm": 0.08494658768177032,
      "learning_rate": 2.6788428734765224e-05,
      "loss": 1.0126953125,
      "step": 80
    },
    {
      "epoch": 3.0,
      "grad_norm": 0.03991111367940903,
      "learning_rate": 2.5722176490809118e-05,
      "loss": 1.009765625,
      "step": 81
    },
    {
      "epoch": 3.037037037037037,
      "grad_norm": 0.041350796818733215,
      "learning_rate": 2.4683696035904928e-05,
      "loss": 1.0244140625,
      "step": 82
    },
    {
      "epoch": 3.074074074074074,
      "grad_norm": 0.04070250317454338,
      "learning_rate": 2.3673972431668306e-05,
      "loss": 0.9990234375,
      "step": 83
    },
    {
      "epoch": 3.111111111111111,
      "grad_norm": 0.03904803469777107,
      "learning_rate": 2.2693963462100117e-05,
      "loss": 1.01220703125,
      "step": 84
    },
    {
      "epoch": 3.148148148148148,
      "grad_norm": 0.038457781076431274,
      "learning_rate": 2.1744598725070347e-05,
      "loss": 1.00048828125,
      "step": 85
    },
    {
      "epoch": 3.185185185185185,
      "grad_norm": 0.03954171761870384,
      "learning_rate": 2.082677875053818e-05,
      "loss": 1.0009765625,
      "step": 86
    },
    {
      "epoch": 3.2222222222222223,
      "grad_norm": 0.03761440888047218,
      "learning_rate": 1.994137414634483e-05,
      "loss": 1.0068359375,
      "step": 87
    },
    {
      "epoch": 3.2222222222222223,
      "eval_loss": 1.04052734375,
      "eval_runtime": 5.5639,
      "eval_samples_per_second": 1.258,
      "eval_steps_per_second": 0.18,
      "step": 87
    },
    {
      "epoch": 3.259259259259259,
      "grad_norm": 0.03947566822171211,
      "learning_rate": 1.9089224772389225e-05,
      "loss": 0.98291015625,
      "step": 88
    },
    {
      "epoch": 3.2962962962962963,
      "grad_norm": 0.04216547682881355,
      "learning_rate": 1.827113894397003e-05,
      "loss": 0.9853515625,
      "step": 89
    },
    {
      "epoch": 3.3333333333333335,
      "grad_norm": 0.042566247284412384,
      "learning_rate": 1.7487892665049627e-05,
      "loss": 0.9873046875,
      "step": 90
    },
    {
      "epoch": 3.3703703703703702,
      "grad_norm": 0.043656375259160995,
      "learning_rate": 1.674022889216737e-05,
      "loss": 0.99365234375,
      "step": 91
    },
    {
      "epoch": 3.4074074074074074,
      "grad_norm": 0.040871571749448776,
      "learning_rate": 1.602885682970026e-05,
      "loss": 0.99267578125,
      "step": 92
    },
    {
      "epoch": 3.4444444444444446,
      "grad_norm": 0.04622000828385353,
      "learning_rate": 1.53544512571397e-05,
      "loss": 0.97998046875,
      "step": 93
    },
    {
      "epoch": 3.4814814814814814,
      "grad_norm": 0.057611219584941864,
      "learning_rate": 1.4717651889022202e-05,
      "loss": 1.048828125,
      "step": 94
    },
    {
      "epoch": 3.5185185185185186,
      "grad_norm": 0.04287968948483467,
      "learning_rate": 1.4119062768121433e-05,
      "loss": 0.98681640625,
      "step": 95
    },
    {
      "epoch": 3.5555555555555554,
      "grad_norm": 0.03957849368453026,
      "learning_rate": 1.3559251692477087e-05,
      "loss": 0.99267578125,
      "step": 96
    },
    {
      "epoch": 3.5925925925925926,
      "grad_norm": 0.040763694792985916,
      "learning_rate": 1.3038749676803994e-05,
      "loss": 0.9853515625,
      "step": 97
    },
    {
      "epoch": 3.6296296296296298,
      "grad_norm": 0.043220002204179764,
      "learning_rate": 1.2558050448792515e-05,
      "loss": 0.96142578125,
      "step": 98
    },
    {
      "epoch": 3.6666666666666665,
      "grad_norm": 0.03983893617987633,
      "learning_rate": 1.2117609980777959e-05,
      "loss": 0.984375,
      "step": 99
    },
    {
      "epoch": 3.7037037037037037,
      "grad_norm": 0.03836297243833542,
      "learning_rate": 1.1717846057223144e-05,
      "loss": 0.97021484375,
      "step": 100
    },
    {
      "epoch": 3.7407407407407405,
      "grad_norm": 0.04044586792588234,
      "learning_rate": 1.1359137878424578e-05,
      "loss": 0.98974609375,
      "step": 101
    },
    {
      "epoch": 3.7777777777777777,
      "grad_norm": 0.04162565618753433,
      "learning_rate": 1.104182570081797e-05,
      "loss": 0.9892578125,
      "step": 102
    },
    {
      "epoch": 3.814814814814815,
      "grad_norm": 0.041506294161081314,
      "learning_rate": 1.076621051422442e-05,
      "loss": 1.0029296875,
      "step": 103
    },
    {
      "epoch": 3.851851851851852,
      "grad_norm": 0.040948495268821716,
      "learning_rate": 1.0532553756343328e-05,
      "loss": 0.990234375,
      "step": 104
    },
    {
      "epoch": 3.888888888888889,
      "grad_norm": 0.03962410241365433,
      "learning_rate": 1.0341077064762893e-05,
      "loss": 0.998046875,
      "step": 105
    },
    {
      "epoch": 3.925925925925926,
      "grad_norm": 0.03852328658103943,
      "learning_rate": 1.019196206672345e-05,
      "loss": 0.9853515625,
      "step": 106
    },
    {
      "epoch": 3.962962962962963,
      "grad_norm": 0.043386559933423996,
      "learning_rate": 1.0085350206833016e-05,
      "loss": 0.99462890625,
      "step": 107
    },
    {
      "epoch": 4.0,
      "grad_norm": 0.04271680861711502,
      "learning_rate": 1.0021342612898534e-05,
      "loss": 0.994140625,
      "step": 108
    },
    {
      "epoch": 4.0,
      "eval_loss": 1.0341796875,
      "eval_runtime": 5.5535,
      "eval_samples_per_second": 1.26,
      "eval_steps_per_second": 0.18,
      "step": 108
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 108,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 0,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 9.389375924459274e+18,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}