{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 0.7246376811594203,
  "eval_steps": 9,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.007246376811594203,
      "grad_norm": 3.457111120223999,
      "learning_rate": 1e-05,
      "loss": 1.9121,
      "step": 1
    },
    {
      "epoch": 0.007246376811594203,
      "eval_loss": 2.3775599002838135,
      "eval_runtime": 26.3985,
      "eval_samples_per_second": 4.432,
      "eval_steps_per_second": 0.568,
      "step": 1
    },
    {
      "epoch": 0.014492753623188406,
      "grad_norm": 3.4186577796936035,
      "learning_rate": 2e-05,
      "loss": 2.4865,
      "step": 2
    },
    {
      "epoch": 0.021739130434782608,
      "grad_norm": 4.268088340759277,
      "learning_rate": 3e-05,
      "loss": 2.3192,
      "step": 3
    },
    {
      "epoch": 0.028985507246376812,
      "grad_norm": 4.2391204833984375,
      "learning_rate": 4e-05,
      "loss": 2.2608,
      "step": 4
    },
    {
      "epoch": 0.036231884057971016,
      "grad_norm": 4.833356857299805,
      "learning_rate": 5e-05,
      "loss": 2.3309,
      "step": 5
    },
    {
      "epoch": 0.043478260869565216,
      "grad_norm": 3.8575387001037598,
      "learning_rate": 6e-05,
      "loss": 1.8617,
      "step": 6
    },
    {
      "epoch": 0.050724637681159424,
      "grad_norm": 3.917534589767456,
      "learning_rate": 7e-05,
      "loss": 1.4952,
      "step": 7
    },
    {
      "epoch": 0.057971014492753624,
      "grad_norm": 3.901942014694214,
      "learning_rate": 8e-05,
      "loss": 1.5037,
      "step": 8
    },
    {
      "epoch": 0.06521739130434782,
      "grad_norm": 3.337735176086426,
      "learning_rate": 9e-05,
      "loss": 1.1529,
      "step": 9
    },
    {
      "epoch": 0.06521739130434782,
      "eval_loss": 0.9570497870445251,
      "eval_runtime": 26.0708,
      "eval_samples_per_second": 4.488,
      "eval_steps_per_second": 0.575,
      "step": 9
    },
    {
      "epoch": 0.07246376811594203,
      "grad_norm": 2.8909802436828613,
      "learning_rate": 0.0001,
      "loss": 0.8145,
      "step": 10
    },
    {
      "epoch": 0.07971014492753623,
      "grad_norm": 3.1760714054107666,
      "learning_rate": 9.99695413509548e-05,
      "loss": 0.6738,
      "step": 11
    },
    {
      "epoch": 0.08695652173913043,
      "grad_norm": 3.3278651237487793,
      "learning_rate": 9.987820251299122e-05,
      "loss": 0.5146,
      "step": 12
    },
    {
      "epoch": 0.09420289855072464,
      "grad_norm": 2.306939125061035,
      "learning_rate": 9.972609476841367e-05,
      "loss": 0.2717,
      "step": 13
    },
    {
      "epoch": 0.10144927536231885,
      "grad_norm": 3.0428788661956787,
      "learning_rate": 9.951340343707852e-05,
      "loss": 0.1852,
      "step": 14
    },
    {
      "epoch": 0.10869565217391304,
      "grad_norm": 4.20550537109375,
      "learning_rate": 9.924038765061042e-05,
      "loss": 0.0783,
      "step": 15
    },
    {
      "epoch": 0.11594202898550725,
      "grad_norm": 1.2375609874725342,
      "learning_rate": 9.890738003669029e-05,
      "loss": 0.0253,
      "step": 16
    },
    {
      "epoch": 0.12318840579710146,
      "grad_norm": 1.1446454524993896,
      "learning_rate": 9.851478631379982e-05,
      "loss": 0.0187,
      "step": 17
    },
    {
      "epoch": 0.13043478260869565,
      "grad_norm": 0.06109399348497391,
      "learning_rate": 9.806308479691595e-05,
      "loss": 0.0009,
      "step": 18
    },
    {
      "epoch": 0.13043478260869565,
      "eval_loss": 0.007476986385881901,
      "eval_runtime": 26.0541,
      "eval_samples_per_second": 4.491,
      "eval_steps_per_second": 0.576,
      "step": 18
    },
    {
      "epoch": 0.13768115942028986,
      "grad_norm": 0.9464291930198669,
      "learning_rate": 9.755282581475769e-05,
      "loss": 0.0033,
      "step": 19
    },
    {
      "epoch": 0.14492753623188406,
      "grad_norm": 0.08613129705190659,
      "learning_rate": 9.698463103929542e-05,
      "loss": 0.0013,
      "step": 20
    },
    {
      "epoch": 0.15217391304347827,
      "grad_norm": 3.3625614643096924,
      "learning_rate": 9.635919272833938e-05,
      "loss": 0.0222,
      "step": 21
    },
    {
      "epoch": 0.15942028985507245,
      "grad_norm": 3.339473009109497,
      "learning_rate": 9.567727288213005e-05,
      "loss": 0.0114,
      "step": 22
    },
    {
      "epoch": 0.16666666666666666,
      "grad_norm": 0.18028295040130615,
      "learning_rate": 9.493970231495835e-05,
      "loss": 0.0018,
      "step": 23
    },
    {
      "epoch": 0.17391304347826086,
      "grad_norm": 1.8303236961364746,
      "learning_rate": 9.414737964294636e-05,
      "loss": 0.0107,
      "step": 24
    },
    {
      "epoch": 0.18115942028985507,
      "grad_norm": 0.08866623044013977,
      "learning_rate": 9.330127018922194e-05,
      "loss": 0.0008,
      "step": 25
    },
    {
      "epoch": 0.18840579710144928,
      "grad_norm": 0.02692384645342827,
      "learning_rate": 9.24024048078213e-05,
      "loss": 0.0004,
      "step": 26
    },
    {
      "epoch": 0.1956521739130435,
      "grad_norm": 0.8676456809043884,
      "learning_rate": 9.145187862775209e-05,
      "loss": 0.0306,
      "step": 27
    },
    {
      "epoch": 0.1956521739130435,
      "eval_loss": 0.004295172169804573,
      "eval_runtime": 26.0556,
      "eval_samples_per_second": 4.49,
      "eval_steps_per_second": 0.576,
      "step": 27
    },
    {
      "epoch": 0.2028985507246377,
      "grad_norm": 1.2788045406341553,
      "learning_rate": 9.045084971874738e-05,
      "loss": 0.0046,
      "step": 28
    },
    {
      "epoch": 0.21014492753623187,
      "grad_norm": 0.012373939156532288,
      "learning_rate": 8.940053768033609e-05,
      "loss": 0.0003,
      "step": 29
    },
    {
      "epoch": 0.21739130434782608,
      "grad_norm": 0.0066170841455459595,
      "learning_rate": 8.83022221559489e-05,
      "loss": 0.0002,
      "step": 30
    },
    {
      "epoch": 0.2246376811594203,
      "grad_norm": 0.3363509476184845,
      "learning_rate": 8.715724127386972e-05,
      "loss": 0.003,
      "step": 31
    },
    {
      "epoch": 0.2318840579710145,
      "grad_norm": 0.8705418705940247,
      "learning_rate": 8.596699001693255e-05,
      "loss": 0.0316,
      "step": 32
    },
    {
      "epoch": 0.2391304347826087,
      "grad_norm": 0.4040496051311493,
      "learning_rate": 8.473291852294987e-05,
      "loss": 0.002,
      "step": 33
    },
    {
      "epoch": 0.2463768115942029,
      "grad_norm": 0.9658280611038208,
      "learning_rate": 8.345653031794292e-05,
      "loss": 0.0149,
      "step": 34
    },
    {
      "epoch": 0.2536231884057971,
      "grad_norm": 0.3180502653121948,
      "learning_rate": 8.213938048432697e-05,
      "loss": 0.0031,
      "step": 35
    },
    {
      "epoch": 0.2608695652173913,
      "grad_norm": 1.5294246673583984,
      "learning_rate": 8.07830737662829e-05,
      "loss": 0.0122,
      "step": 36
    },
    {
      "epoch": 0.2608695652173913,
      "eval_loss": 0.003868628991767764,
      "eval_runtime": 26.0384,
      "eval_samples_per_second": 4.493,
      "eval_steps_per_second": 0.576,
      "step": 36
    },
    {
      "epoch": 0.26811594202898553,
      "grad_norm": 0.48853886127471924,
      "learning_rate": 7.938926261462366e-05,
      "loss": 0.017,
      "step": 37
    },
    {
      "epoch": 0.2753623188405797,
      "grad_norm": 0.01280419621616602,
      "learning_rate": 7.795964517353735e-05,
      "loss": 0.0003,
      "step": 38
    },
    {
      "epoch": 0.2826086956521739,
      "grad_norm": 0.045505233108997345,
      "learning_rate": 7.649596321166024e-05,
      "loss": 0.0008,
      "step": 39
    },
    {
      "epoch": 0.2898550724637681,
      "grad_norm": 0.007184368558228016,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0002,
      "step": 40
    },
    {
      "epoch": 0.2971014492753623,
      "grad_norm": 0.03741440922021866,
      "learning_rate": 7.347357813929454e-05,
      "loss": 0.0004,
      "step": 41
    },
    {
      "epoch": 0.30434782608695654,
      "grad_norm": 0.1868954449892044,
      "learning_rate": 7.191855733945387e-05,
      "loss": 0.0022,
      "step": 42
    },
    {
      "epoch": 0.3115942028985507,
      "grad_norm": 0.20426076650619507,
      "learning_rate": 7.033683215379002e-05,
      "loss": 0.0023,
      "step": 43
    },
    {
      "epoch": 0.3188405797101449,
      "grad_norm": 0.03536685183644295,
      "learning_rate": 6.873032967079561e-05,
      "loss": 0.0005,
      "step": 44
    },
    {
      "epoch": 0.32608695652173914,
      "grad_norm": 0.5953764319419861,
      "learning_rate": 6.710100716628344e-05,
      "loss": 0.0033,
      "step": 45
    },
    {
      "epoch": 0.32608695652173914,
      "eval_loss": 0.002579378429800272,
      "eval_runtime": 26.0467,
      "eval_samples_per_second": 4.492,
      "eval_steps_per_second": 0.576,
      "step": 45
    },
    {
      "epoch": 0.3333333333333333,
      "grad_norm": 0.03366972506046295,
      "learning_rate": 6.545084971874738e-05,
      "loss": 0.0007,
      "step": 46
    },
    {
      "epoch": 0.34057971014492755,
      "grad_norm": 2.9528863430023193,
      "learning_rate": 6.378186779084995e-05,
      "loss": 0.0285,
      "step": 47
    },
    {
      "epoch": 0.34782608695652173,
      "grad_norm": 0.3075985908508301,
      "learning_rate": 6.209609477998338e-05,
      "loss": 0.0033,
      "step": 48
    },
    {
      "epoch": 0.35507246376811596,
      "grad_norm": 0.015260318294167519,
      "learning_rate": 6.0395584540887963e-05,
      "loss": 0.0003,
      "step": 49
    },
    {
      "epoch": 0.36231884057971014,
      "grad_norm": 0.02092459797859192,
      "learning_rate": 5.868240888334653e-05,
      "loss": 0.0003,
      "step": 50
    },
    {
      "epoch": 0.3695652173913043,
      "grad_norm": 0.18774715065956116,
      "learning_rate": 5.695865504800327e-05,
      "loss": 0.0034,
      "step": 51
    },
    {
      "epoch": 0.37681159420289856,
      "grad_norm": 0.08688025176525116,
      "learning_rate": 5.522642316338268e-05,
      "loss": 0.0008,
      "step": 52
    },
    {
      "epoch": 0.38405797101449274,
      "grad_norm": 0.08076933026313782,
      "learning_rate": 5.348782368720626e-05,
      "loss": 0.0009,
      "step": 53
    },
    {
      "epoch": 0.391304347826087,
      "grad_norm": 3.284877061843872,
      "learning_rate": 5.174497483512506e-05,
      "loss": 0.0604,
      "step": 54
    },
    {
      "epoch": 0.391304347826087,
      "eval_loss": 0.0013279261766001582,
      "eval_runtime": 26.0413,
      "eval_samples_per_second": 4.493,
      "eval_steps_per_second": 0.576,
      "step": 54
    },
    {
      "epoch": 0.39855072463768115,
      "grad_norm": 0.06980286538600922,
      "learning_rate": 5e-05,
      "loss": 0.0008,
      "step": 55
    },
    {
      "epoch": 0.4057971014492754,
      "grad_norm": 0.11990461498498917,
      "learning_rate": 4.825502516487497e-05,
      "loss": 0.0014,
      "step": 56
    },
    {
      "epoch": 0.41304347826086957,
      "grad_norm": 0.1687093824148178,
      "learning_rate": 4.6512176312793736e-05,
      "loss": 0.0011,
      "step": 57
    },
    {
      "epoch": 0.42028985507246375,
      "grad_norm": 0.02264598198235035,
      "learning_rate": 4.477357683661734e-05,
      "loss": 0.0004,
      "step": 58
    },
    {
      "epoch": 0.427536231884058,
      "grad_norm": 0.030011940747499466,
      "learning_rate": 4.3041344951996746e-05,
      "loss": 0.0005,
      "step": 59
    },
    {
      "epoch": 0.43478260869565216,
      "grad_norm": 0.04578249529004097,
      "learning_rate": 4.131759111665349e-05,
      "loss": 0.0006,
      "step": 60
    },
    {
      "epoch": 0.4420289855072464,
      "grad_norm": 0.38091355562210083,
      "learning_rate": 3.960441545911204e-05,
      "loss": 0.0058,
      "step": 61
    },
    {
      "epoch": 0.4492753623188406,
      "grad_norm": 0.14858709275722504,
      "learning_rate": 3.790390522001662e-05,
      "loss": 0.0006,
      "step": 62
    },
    {
      "epoch": 0.45652173913043476,
      "grad_norm": 0.03918566182255745,
      "learning_rate": 3.6218132209150045e-05,
      "loss": 0.0004,
      "step": 63
    },
    {
      "epoch": 0.45652173913043476,
      "eval_loss": 0.0008458091178908944,
      "eval_runtime": 26.0471,
      "eval_samples_per_second": 4.492,
      "eval_steps_per_second": 0.576,
      "step": 63
    },
    {
      "epoch": 0.463768115942029,
      "grad_norm": 0.010256417095661163,
      "learning_rate": 3.4549150281252636e-05,
      "loss": 0.0002,
      "step": 64
    },
    {
      "epoch": 0.47101449275362317,
      "grad_norm": 0.11472321301698685,
      "learning_rate": 3.289899283371657e-05,
      "loss": 0.0007,
      "step": 65
    },
    {
      "epoch": 0.4782608695652174,
      "grad_norm": 0.01833566464483738,
      "learning_rate": 3.12696703292044e-05,
      "loss": 0.0004,
      "step": 66
    },
    {
      "epoch": 0.4855072463768116,
      "grad_norm": 0.02492792159318924,
      "learning_rate": 2.9663167846209998e-05,
      "loss": 0.0005,
      "step": 67
    },
    {
      "epoch": 0.4927536231884058,
      "grad_norm": 0.024905331432819366,
      "learning_rate": 2.8081442660546125e-05,
      "loss": 0.0004,
      "step": 68
    },
    {
      "epoch": 0.5,
      "grad_norm": 0.0337512344121933,
      "learning_rate": 2.6526421860705473e-05,
      "loss": 0.0004,
      "step": 69
    },
    {
      "epoch": 0.5072463768115942,
      "grad_norm": 0.003508348483592272,
      "learning_rate": 2.500000000000001e-05,
      "loss": 0.0001,
      "step": 70
    },
    {
      "epoch": 0.5144927536231884,
      "grad_norm": 0.13854865729808807,
      "learning_rate": 2.350403678833976e-05,
      "loss": 0.0017,
      "step": 71
    },
    {
      "epoch": 0.5217391304347826,
      "grad_norm": 0.05484834685921669,
      "learning_rate": 2.2040354826462668e-05,
      "loss": 0.0004,
      "step": 72
    },
    {
      "epoch": 0.5217391304347826,
      "eval_loss": 0.0004127143183723092,
      "eval_runtime": 26.0422,
      "eval_samples_per_second": 4.493,
      "eval_steps_per_second": 0.576,
      "step": 72
    },
    {
      "epoch": 0.5289855072463768,
      "grad_norm": 0.006606305483728647,
      "learning_rate": 2.061073738537635e-05,
      "loss": 0.0001,
      "step": 73
    },
    {
      "epoch": 0.5362318840579711,
      "grad_norm": 0.0063642412424087524,
      "learning_rate": 1.9216926233717085e-05,
      "loss": 0.0001,
      "step": 74
    },
    {
      "epoch": 0.5434782608695652,
      "grad_norm": 0.016659699380397797,
      "learning_rate": 1.7860619515673033e-05,
      "loss": 0.0003,
      "step": 75
    },
    {
      "epoch": 0.5507246376811594,
      "grad_norm": 0.002692802343517542,
      "learning_rate": 1.6543469682057106e-05,
      "loss": 0.0001,
      "step": 76
    },
    {
      "epoch": 0.5579710144927537,
      "grad_norm": 0.02030947059392929,
      "learning_rate": 1.526708147705013e-05,
      "loss": 0.0003,
      "step": 77
    },
    {
      "epoch": 0.5652173913043478,
      "grad_norm": 0.004038242623209953,
      "learning_rate": 1.4033009983067452e-05,
      "loss": 0.0001,
      "step": 78
    },
    {
      "epoch": 0.572463768115942,
      "grad_norm": 0.004218004643917084,
      "learning_rate": 1.2842758726130283e-05,
      "loss": 0.0001,
      "step": 79
    },
    {
      "epoch": 0.5797101449275363,
      "grad_norm": 0.0025967604015022516,
      "learning_rate": 1.1697777844051105e-05,
      "loss": 0.0001,
      "step": 80
    },
    {
      "epoch": 0.5869565217391305,
      "grad_norm": 0.04226625710725784,
      "learning_rate": 1.0599462319663905e-05,
      "loss": 0.0004,
      "step": 81
    },
    {
      "epoch": 0.5869565217391305,
      "eval_loss": 0.000350808521034196,
      "eval_runtime": 26.0659,
      "eval_samples_per_second": 4.489,
      "eval_steps_per_second": 0.575,
      "step": 81
    },
    {
      "epoch": 0.5942028985507246,
      "grad_norm": 0.0268058180809021,
      "learning_rate": 9.549150281252633e-06,
      "loss": 0.0005,
      "step": 82
    },
    {
      "epoch": 0.6014492753623188,
      "grad_norm": 0.004873110447078943,
      "learning_rate": 8.548121372247918e-06,
      "loss": 0.0001,
      "step": 83
    },
    {
      "epoch": 0.6086956521739131,
      "grad_norm": 0.00749048450961709,
      "learning_rate": 7.597595192178702e-06,
      "loss": 0.0002,
      "step": 84
    },
    {
      "epoch": 0.6159420289855072,
      "grad_norm": 0.004632422234863043,
      "learning_rate": 6.698729810778065e-06,
      "loss": 0.0001,
      "step": 85
    },
    {
      "epoch": 0.6231884057971014,
      "grad_norm": 0.15520110726356506,
      "learning_rate": 5.852620357053651e-06,
      "loss": 0.0021,
      "step": 86
    },
    {
      "epoch": 0.6304347826086957,
      "grad_norm": 0.009636754170060158,
      "learning_rate": 5.060297685041659e-06,
      "loss": 0.0002,
      "step": 87
    },
    {
      "epoch": 0.6376811594202898,
      "grad_norm": 0.29421183466911316,
      "learning_rate": 4.322727117869951e-06,
      "loss": 0.0013,
      "step": 88
    },
    {
      "epoch": 0.644927536231884,
      "grad_norm": 0.002441299380734563,
      "learning_rate": 3.6408072716606346e-06,
      "loss": 0.0001,
      "step": 89
    },
    {
      "epoch": 0.6521739130434783,
      "grad_norm": 0.0820772722363472,
      "learning_rate": 3.0153689607045845e-06,
      "loss": 0.0011,
      "step": 90
    },
    {
      "epoch": 0.6521739130434783,
      "eval_loss": 0.0003161544445902109,
      "eval_runtime": 26.0467,
      "eval_samples_per_second": 4.492,
      "eval_steps_per_second": 0.576,
      "step": 90
    },
    {
      "epoch": 0.6594202898550725,
      "grad_norm": 0.0031900727190077305,
      "learning_rate": 2.4471741852423237e-06,
      "loss": 0.0001,
      "step": 91
    },
    {
      "epoch": 0.6666666666666666,
      "grad_norm": 0.00806911289691925,
      "learning_rate": 1.9369152030840556e-06,
      "loss": 0.0002,
      "step": 92
    },
    {
      "epoch": 0.6739130434782609,
      "grad_norm": 0.0062135085463523865,
      "learning_rate": 1.4852136862001764e-06,
      "loss": 0.0001,
      "step": 93
    },
    {
      "epoch": 0.6811594202898551,
      "grad_norm": 0.0019510157871991396,
      "learning_rate": 1.0926199633097157e-06,
      "loss": 0.0001,
      "step": 94
    },
    {
      "epoch": 0.6884057971014492,
      "grad_norm": 0.01378517597913742,
      "learning_rate": 7.596123493895991e-07,
      "loss": 0.0002,
      "step": 95
    },
    {
      "epoch": 0.6956521739130435,
      "grad_norm": 0.11113560199737549,
      "learning_rate": 4.865965629214819e-07,
      "loss": 0.0008,
      "step": 96
    },
    {
      "epoch": 0.7028985507246377,
      "grad_norm": 0.0036437332164496183,
      "learning_rate": 2.7390523158633554e-07,
      "loss": 0.0001,
      "step": 97
    },
    {
      "epoch": 0.7101449275362319,
      "grad_norm": 0.12289454787969589,
      "learning_rate": 1.2179748700879012e-07,
      "loss": 0.0011,
      "step": 98
    },
    {
      "epoch": 0.717391304347826,
      "grad_norm": 0.011756442487239838,
      "learning_rate": 3.04586490452119e-08,
      "loss": 0.0003,
      "step": 99
    },
    {
      "epoch": 0.717391304347826,
      "eval_loss": 0.00032029286376200616,
      "eval_runtime": 26.0531,
      "eval_samples_per_second": 4.491,
      "eval_steps_per_second": 0.576,
      "step": 99
    },
    {
      "epoch": 0.7246376811594203,
      "grad_norm": 0.008075772784650326,
      "learning_rate": 0.0,
      "loss": 0.0002,
      "step": 100
    }
  ],
  "logging_steps": 1,
  "max_steps": 100,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 25,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.483774567120896e+17,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}