{
  "best_metric": 0.9318181818181818,
  "best_model_checkpoint": "beit-base-patch16-224-85-fold5/checkpoint-92",
  "epoch": 100.0,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.29545454545454547,
      "eval_loss": 1.089578628540039,
      "eval_runtime": 0.7021,
      "eval_samples_per_second": 62.667,
      "eval_steps_per_second": 2.849,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7272727272727273,
      "eval_loss": 0.6455699801445007,
      "eval_runtime": 0.6856,
      "eval_samples_per_second": 64.174,
      "eval_steps_per_second": 2.917,
      "step": 4
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7045454545454546,
      "eval_loss": 1.0355168581008911,
      "eval_runtime": 0.6996,
      "eval_samples_per_second": 62.897,
      "eval_steps_per_second": 2.859,
      "step": 6
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7045454545454546,
      "eval_loss": 0.9123974442481995,
      "eval_runtime": 0.6813,
      "eval_samples_per_second": 64.582,
      "eval_steps_per_second": 2.936,
      "step": 8
    },
    {
      "epoch": 5.0,
      "grad_norm": 9.428820610046387,
      "learning_rate": 2.5e-05,
      "loss": 0.7607,
      "step": 10
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7954545454545454,
      "eval_loss": 0.5808687210083008,
      "eval_runtime": 0.672,
      "eval_samples_per_second": 65.48,
      "eval_steps_per_second": 2.976,
      "step": 10
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.6812124848365784,
      "eval_runtime": 0.6683,
      "eval_samples_per_second": 65.834,
      "eval_steps_per_second": 2.992,
      "step": 12
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.6529408097267151,
      "eval_runtime": 0.6753,
      "eval_samples_per_second": 65.16,
      "eval_steps_per_second": 2.962,
      "step": 14
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.7272727272727273,
      "eval_loss": 0.717404305934906,
      "eval_runtime": 0.7009,
      "eval_samples_per_second": 62.778,
      "eval_steps_per_second": 2.854,
      "step": 16
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.6136363636363636,
      "eval_loss": 0.6618563532829285,
      "eval_runtime": 0.6832,
      "eval_samples_per_second": 64.403,
      "eval_steps_per_second": 2.927,
      "step": 18
    },
    {
      "epoch": 10.0,
      "grad_norm": 6.730888843536377,
      "learning_rate": 5e-05,
      "loss": 0.4221,
      "step": 20
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.80633944272995,
      "eval_runtime": 0.6867,
      "eval_samples_per_second": 64.075,
      "eval_steps_per_second": 2.912,
      "step": 20
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.6371808648109436,
      "eval_runtime": 0.6802,
      "eval_samples_per_second": 64.689,
      "eval_steps_per_second": 2.94,
      "step": 22
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.588646411895752,
      "eval_runtime": 0.6871,
      "eval_samples_per_second": 64.039,
      "eval_steps_per_second": 2.911,
      "step": 24
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.6363636363636364,
      "eval_loss": 0.6359339356422424,
      "eval_runtime": 0.6771,
      "eval_samples_per_second": 64.986,
      "eval_steps_per_second": 2.954,
      "step": 26
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.5584580898284912,
      "eval_runtime": 0.6833,
      "eval_samples_per_second": 64.391,
      "eval_steps_per_second": 2.927,
      "step": 28
    },
    {
      "epoch": 15.0,
      "grad_norm": 9.09820556640625,
      "learning_rate": 4.722222222222222e-05,
      "loss": 0.3287,
      "step": 30
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.7272727272727273,
      "eval_loss": 0.45406511425971985,
      "eval_runtime": 0.6778,
      "eval_samples_per_second": 64.917,
      "eval_steps_per_second": 2.951,
      "step": 30
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5681818181818182,
      "eval_loss": 0.7624431252479553,
      "eval_runtime": 0.6791,
      "eval_samples_per_second": 64.788,
      "eval_steps_per_second": 2.945,
      "step": 32
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.6805771589279175,
      "eval_runtime": 0.6896,
      "eval_samples_per_second": 63.807,
      "eval_steps_per_second": 2.9,
      "step": 34
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.7272727272727273,
      "eval_loss": 0.7707971930503845,
      "eval_runtime": 0.6822,
      "eval_samples_per_second": 64.496,
      "eval_steps_per_second": 2.932,
      "step": 36
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.7272727272727273,
      "eval_loss": 0.416960209608078,
      "eval_runtime": 0.6839,
      "eval_samples_per_second": 64.337,
      "eval_steps_per_second": 2.924,
      "step": 38
    },
    {
      "epoch": 20.0,
      "grad_norm": 12.929576873779297,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.3663,
      "step": 40
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7727272727272727,
      "eval_loss": 0.42817583680152893,
      "eval_runtime": 0.684,
      "eval_samples_per_second": 64.327,
      "eval_steps_per_second": 2.924,
      "step": 40
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.5613061785697937,
      "eval_runtime": 0.6869,
      "eval_samples_per_second": 64.059,
      "eval_steps_per_second": 2.912,
      "step": 42
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.4025155007839203,
      "eval_runtime": 0.7081,
      "eval_samples_per_second": 62.136,
      "eval_steps_per_second": 2.824,
      "step": 44
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.7954545454545454,
      "eval_loss": 0.41085565090179443,
      "eval_runtime": 0.7009,
      "eval_samples_per_second": 62.774,
      "eval_steps_per_second": 2.853,
      "step": 46
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.4372767508029938,
      "eval_runtime": 0.6888,
      "eval_samples_per_second": 63.878,
      "eval_steps_per_second": 2.904,
      "step": 48
    },
    {
      "epoch": 25.0,
      "grad_norm": 6.6011061668396,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.2344,
      "step": 50
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3210701644420624,
      "eval_runtime": 0.6782,
      "eval_samples_per_second": 64.873,
      "eval_steps_per_second": 2.949,
      "step": 50
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.5560834407806396,
      "eval_runtime": 0.6873,
      "eval_samples_per_second": 64.014,
      "eval_steps_per_second": 2.91,
      "step": 52
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.31488853693008423,
      "eval_runtime": 0.6879,
      "eval_samples_per_second": 63.96,
      "eval_steps_per_second": 2.907,
      "step": 54
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.7954545454545454,
      "eval_loss": 0.31661316752433777,
      "eval_runtime": 0.679,
      "eval_samples_per_second": 64.805,
      "eval_steps_per_second": 2.946,
      "step": 56
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.4163520932197571,
      "eval_runtime": 0.6778,
      "eval_samples_per_second": 64.92,
      "eval_steps_per_second": 2.951,
      "step": 58
    },
    {
      "epoch": 30.0,
      "grad_norm": 5.3891143798828125,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.2051,
      "step": 60
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.4345444440841675,
      "eval_runtime": 0.6869,
      "eval_samples_per_second": 64.054,
      "eval_steps_per_second": 2.912,
      "step": 60
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.31803134083747864,
      "eval_runtime": 0.6887,
      "eval_samples_per_second": 63.886,
      "eval_steps_per_second": 2.904,
      "step": 62
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.3673260509967804,
      "eval_runtime": 0.6835,
      "eval_samples_per_second": 64.377,
      "eval_steps_per_second": 2.926,
      "step": 64
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.43128955364227295,
      "eval_runtime": 0.682,
      "eval_samples_per_second": 64.517,
      "eval_steps_per_second": 2.933,
      "step": 66
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.43590399622917175,
      "eval_runtime": 0.6788,
      "eval_samples_per_second": 64.824,
      "eval_steps_per_second": 2.947,
      "step": 68
    },
    {
      "epoch": 35.0,
      "grad_norm": 3.639585256576538,
      "learning_rate": 3.611111111111111e-05,
      "loss": 0.1694,
      "step": 70
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.36998674273490906,
      "eval_runtime": 0.6829,
      "eval_samples_per_second": 64.428,
      "eval_steps_per_second": 2.929,
      "step": 70
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.7954545454545454,
      "eval_loss": 0.584349513053894,
      "eval_runtime": 0.6937,
      "eval_samples_per_second": 63.427,
      "eval_steps_per_second": 2.883,
      "step": 72
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.4064345061779022,
      "eval_runtime": 0.7057,
      "eval_samples_per_second": 62.346,
      "eval_steps_per_second": 2.834,
      "step": 74
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.39915817975997925,
      "eval_runtime": 0.682,
      "eval_samples_per_second": 64.516,
      "eval_steps_per_second": 2.933,
      "step": 76
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.3152655363082886,
      "eval_runtime": 0.6899,
      "eval_samples_per_second": 63.778,
      "eval_steps_per_second": 2.899,
      "step": 78
    },
    {
      "epoch": 40.0,
      "grad_norm": 5.693894863128662,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1566,
      "step": 80
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.5581021904945374,
      "eval_runtime": 0.6812,
      "eval_samples_per_second": 64.596,
      "eval_steps_per_second": 2.936,
      "step": 80
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.2921257019042969,
      "eval_runtime": 0.6885,
      "eval_samples_per_second": 63.908,
      "eval_steps_per_second": 2.905,
      "step": 82
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.32171034812927246,
      "eval_runtime": 0.684,
      "eval_samples_per_second": 64.326,
      "eval_steps_per_second": 2.924,
      "step": 84
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.32552286982536316,
      "eval_runtime": 0.6845,
      "eval_samples_per_second": 64.277,
      "eval_steps_per_second": 2.922,
      "step": 86
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.75,
      "eval_loss": 0.723772406578064,
      "eval_runtime": 0.7032,
      "eval_samples_per_second": 62.569,
      "eval_steps_per_second": 2.844,
      "step": 88
    },
    {
      "epoch": 45.0,
      "grad_norm": 9.99726676940918,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.1389,
      "step": 90
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.40532952547073364,
      "eval_runtime": 0.6753,
      "eval_samples_per_second": 65.154,
      "eval_steps_per_second": 2.962,
      "step": 90
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.9318181818181818,
      "eval_loss": 0.24986252188682556,
      "eval_runtime": 0.6918,
      "eval_samples_per_second": 63.603,
      "eval_steps_per_second": 2.891,
      "step": 92
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.2584282159805298,
      "eval_runtime": 0.6836,
      "eval_samples_per_second": 64.362,
      "eval_steps_per_second": 2.926,
      "step": 94
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.44320380687713623,
      "eval_runtime": 0.6851,
      "eval_samples_per_second": 64.225,
      "eval_steps_per_second": 2.919,
      "step": 96
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.7954545454545454,
      "eval_loss": 0.6965329647064209,
      "eval_runtime": 0.6828,
      "eval_samples_per_second": 64.443,
      "eval_steps_per_second": 2.929,
      "step": 98
    },
    {
      "epoch": 50.0,
      "grad_norm": 5.456106662750244,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1311,
      "step": 100
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.39098644256591797,
      "eval_runtime": 0.6828,
      "eval_samples_per_second": 64.437,
      "eval_steps_per_second": 2.929,
      "step": 100
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.30165335536003113,
      "eval_runtime": 0.6957,
      "eval_samples_per_second": 63.242,
      "eval_steps_per_second": 2.875,
      "step": 102
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.30502772331237793,
      "eval_runtime": 0.6903,
      "eval_samples_per_second": 63.743,
      "eval_steps_per_second": 2.897,
      "step": 104
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.21932415664196014,
      "eval_runtime": 0.6995,
      "eval_samples_per_second": 62.902,
      "eval_steps_per_second": 2.859,
      "step": 106
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.23692327737808228,
      "eval_runtime": 0.6924,
      "eval_samples_per_second": 63.544,
      "eval_steps_per_second": 2.888,
      "step": 108
    },
    {
      "epoch": 55.0,
      "grad_norm": 2.8878471851348877,
      "learning_rate": 2.5e-05,
      "loss": 0.1386,
      "step": 110
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.31426045298576355,
      "eval_runtime": 0.6968,
      "eval_samples_per_second": 63.146,
      "eval_steps_per_second": 2.87,
      "step": 110
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.29323482513427734,
      "eval_runtime": 0.6884,
      "eval_samples_per_second": 63.916,
      "eval_steps_per_second": 2.905,
      "step": 112
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.27247121930122375,
      "eval_runtime": 0.6889,
      "eval_samples_per_second": 63.868,
      "eval_steps_per_second": 2.903,
      "step": 114
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.566352128982544,
      "eval_runtime": 0.6829,
      "eval_samples_per_second": 64.432,
      "eval_steps_per_second": 2.929,
      "step": 116
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.5874523520469666,
      "eval_runtime": 0.6823,
      "eval_samples_per_second": 64.484,
      "eval_steps_per_second": 2.931,
      "step": 118
    },
    {
      "epoch": 60.0,
      "grad_norm": 5.098616123199463,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1194,
      "step": 120
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.46228501200675964,
      "eval_runtime": 0.6928,
      "eval_samples_per_second": 63.507,
      "eval_steps_per_second": 2.887,
      "step": 120
    },
    {
      "epoch": 61.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.4716045558452606,
      "eval_runtime": 0.6856,
      "eval_samples_per_second": 64.174,
      "eval_steps_per_second": 2.917,
      "step": 122
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.5028097033500671,
      "eval_runtime": 0.6887,
      "eval_samples_per_second": 63.887,
      "eval_steps_per_second": 2.904,
      "step": 124
    },
    {
      "epoch": 63.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.4557519257068634,
      "eval_runtime": 0.6871,
      "eval_samples_per_second": 64.04,
      "eval_steps_per_second": 2.911,
      "step": 126
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.47977954149246216,
      "eval_runtime": 0.6832,
      "eval_samples_per_second": 64.4,
      "eval_steps_per_second": 2.927,
      "step": 128
    },
    {
      "epoch": 65.0,
      "grad_norm": 4.329418659210205,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.1122,
      "step": 130
    },
    {
      "epoch": 65.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.3826620876789093,
      "eval_runtime": 0.6842,
      "eval_samples_per_second": 64.31,
      "eval_steps_per_second": 2.923,
      "step": 130
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.3652952313423157,
      "eval_runtime": 0.6915,
      "eval_samples_per_second": 63.63,
      "eval_steps_per_second": 2.892,
      "step": 132
    },
    {
      "epoch": 67.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.3972075581550598,
      "eval_runtime": 0.6803,
      "eval_samples_per_second": 64.678,
      "eval_steps_per_second": 2.94,
      "step": 134
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.7727272727272727,
      "eval_loss": 0.5704622864723206,
      "eval_runtime": 0.6889,
      "eval_samples_per_second": 63.868,
      "eval_steps_per_second": 2.903,
      "step": 136
    },
    {
      "epoch": 69.0,
      "eval_accuracy": 0.7727272727272727,
      "eval_loss": 0.5934601426124573,
      "eval_runtime": 0.6795,
      "eval_samples_per_second": 64.752,
      "eval_steps_per_second": 2.943,
      "step": 138
    },
    {
      "epoch": 70.0,
      "grad_norm": 3.6549811363220215,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1041,
      "step": 140
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.3905385136604309,
      "eval_runtime": 0.6809,
      "eval_samples_per_second": 64.623,
      "eval_steps_per_second": 2.937,
      "step": 140
    },
    {
      "epoch": 71.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.27905356884002686,
      "eval_runtime": 0.6869,
      "eval_samples_per_second": 64.051,
      "eval_steps_per_second": 2.911,
      "step": 142
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.28452473878860474,
      "eval_runtime": 0.6826,
      "eval_samples_per_second": 64.463,
      "eval_steps_per_second": 2.93,
      "step": 144
    },
    {
      "epoch": 73.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.24005986750125885,
      "eval_runtime": 0.6824,
      "eval_samples_per_second": 64.477,
      "eval_steps_per_second": 2.931,
      "step": 146
    },
    {
      "epoch": 74.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.2259605973958969,
      "eval_runtime": 0.6828,
      "eval_samples_per_second": 64.437,
      "eval_steps_per_second": 2.929,
      "step": 148
    },
    {
      "epoch": 75.0,
      "grad_norm": 3.5433380603790283,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.0982,
      "step": 150
    },
    {
      "epoch": 75.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.2454221248626709,
      "eval_runtime": 0.6812,
      "eval_samples_per_second": 64.589,
      "eval_steps_per_second": 2.936,
      "step": 150
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3772827386856079,
      "eval_runtime": 0.7099,
      "eval_samples_per_second": 61.983,
      "eval_steps_per_second": 2.817,
      "step": 152
    },
    {
      "epoch": 77.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.618475615978241,
      "eval_runtime": 0.6861,
      "eval_samples_per_second": 64.132,
      "eval_steps_per_second": 2.915,
      "step": 154
    },
    {
      "epoch": 78.0,
      "eval_accuracy": 0.7727272727272727,
      "eval_loss": 0.7238039374351501,
      "eval_runtime": 0.6796,
      "eval_samples_per_second": 64.746,
      "eval_steps_per_second": 2.943,
      "step": 156
    },
    {
      "epoch": 79.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.5469216704368591,
      "eval_runtime": 0.687,
      "eval_samples_per_second": 64.051,
      "eval_steps_per_second": 2.911,
      "step": 158
    },
    {
      "epoch": 80.0,
      "grad_norm": 3.139439344406128,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1065,
      "step": 160
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.4317708909511566,
      "eval_runtime": 0.6729,
      "eval_samples_per_second": 65.391,
      "eval_steps_per_second": 2.972,
      "step": 160
    },
    {
      "epoch": 81.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.33475610613822937,
      "eval_runtime": 0.685,
      "eval_samples_per_second": 64.237,
      "eval_steps_per_second": 2.92,
      "step": 162
    },
    {
      "epoch": 82.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.3040803372859955,
      "eval_runtime": 0.7242,
      "eval_samples_per_second": 60.758,
      "eval_steps_per_second": 2.762,
      "step": 164
    },
    {
      "epoch": 83.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.33496472239494324,
      "eval_runtime": 0.6781,
      "eval_samples_per_second": 64.888,
      "eval_steps_per_second": 2.949,
      "step": 166
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.34641969203948975,
      "eval_runtime": 0.6832,
      "eval_samples_per_second": 64.4,
      "eval_steps_per_second": 2.927,
      "step": 168
    },
    {
      "epoch": 85.0,
      "grad_norm": 3.164541482925415,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.0829,
      "step": 170
    },
    {
      "epoch": 85.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.33745819330215454,
      "eval_runtime": 0.6769,
      "eval_samples_per_second": 64.998,
      "eval_steps_per_second": 2.954,
      "step": 170
    },
    {
      "epoch": 86.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.33091866970062256,
      "eval_runtime": 0.6868,
      "eval_samples_per_second": 64.069,
      "eval_steps_per_second": 2.912,
      "step": 172
    },
    {
      "epoch": 87.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.33253809809684753,
      "eval_runtime": 0.6828,
      "eval_samples_per_second": 64.44,
      "eval_steps_per_second": 2.929,
      "step": 174
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3441101312637329,
      "eval_runtime": 0.6847,
      "eval_samples_per_second": 64.259,
      "eval_steps_per_second": 2.921,
      "step": 176
    },
    {
      "epoch": 89.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.34555572271347046,
      "eval_runtime": 0.7006,
      "eval_samples_per_second": 62.805,
      "eval_steps_per_second": 2.855,
      "step": 178
    },
    {
      "epoch": 90.0,
      "grad_norm": 2.9145150184631348,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0902,
      "step": 180
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.32443130016326904,
      "eval_runtime": 0.6801,
      "eval_samples_per_second": 64.698,
      "eval_steps_per_second": 2.941,
      "step": 180
    },
    {
      "epoch": 91.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.3126292824745178,
      "eval_runtime": 0.6994,
      "eval_samples_per_second": 62.913,
      "eval_steps_per_second": 2.86,
      "step": 182
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.3116739094257355,
      "eval_runtime": 0.6939,
      "eval_samples_per_second": 63.409,
      "eval_steps_per_second": 2.882,
      "step": 184
    },
    {
      "epoch": 93.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.28766298294067383,
      "eval_runtime": 0.6892,
      "eval_samples_per_second": 63.845,
      "eval_steps_per_second": 2.902,
      "step": 186
    },
    {
      "epoch": 94.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.26428502798080444,
      "eval_runtime": 0.7034,
      "eval_samples_per_second": 62.55,
      "eval_steps_per_second": 2.843,
      "step": 188
    },
    {
      "epoch": 95.0,
      "grad_norm": 2.80922269821167,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.0838,
      "step": 190
    },
    {
      "epoch": 95.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.25247523188591003,
      "eval_runtime": 0.6832,
      "eval_samples_per_second": 64.4,
      "eval_steps_per_second": 2.927,
      "step": 190
    },
    {
      "epoch": 96.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.24620772898197174,
      "eval_runtime": 0.6948,
      "eval_samples_per_second": 63.325,
      "eval_steps_per_second": 2.878,
      "step": 192
    },
    {
      "epoch": 97.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.24165372550487518,
      "eval_runtime": 0.6884,
      "eval_samples_per_second": 63.918,
      "eval_steps_per_second": 2.905,
      "step": 194
    },
    {
      "epoch": 98.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.24015091359615326,
      "eval_runtime": 0.6834,
      "eval_samples_per_second": 64.382,
      "eval_steps_per_second": 2.926,
      "step": 196
    },
    {
      "epoch": 99.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.24094568192958832,
      "eval_runtime": 0.6957,
      "eval_samples_per_second": 63.249,
      "eval_steps_per_second": 2.875,
      "step": 198
    },
    {
      "epoch": 100.0,
      "grad_norm": 2.4933438301086426,
      "learning_rate": 0.0,
      "loss": 0.0747,
      "step": 200
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.24256624281406403,
      "eval_runtime": 0.6904,
      "eval_samples_per_second": 63.734,
      "eval_steps_per_second": 2.897,
      "step": 200
    },
    {
      "epoch": 100.0,
      "step": 200,
      "total_flos": 1.8822505078960128e+18,
      "train_loss": 0.19619473487138747,
      "train_runtime": 1209.7214,
      "train_samples_per_second": 20.087,
      "train_steps_per_second": 0.165
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.9318181818181818,
      "eval_loss": 0.24986252188682556,
      "eval_runtime": 0.7488,
      "eval_samples_per_second": 58.758,
      "eval_steps_per_second": 2.671,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 1.8822505078960128e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}