{
  "best_metric": 0.9318181818181818,
  "best_model_checkpoint": "beit-base-patch16-224-85-fold2/checkpoint-158",
  "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": 0.7272727272727273,
      "eval_loss": 0.6056814193725586,
      "eval_runtime": 0.7066,
      "eval_samples_per_second": 62.267,
      "eval_steps_per_second": 2.83,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7045454545454546,
      "eval_loss": 0.663891077041626,
      "eval_runtime": 0.7288,
      "eval_samples_per_second": 60.373,
      "eval_steps_per_second": 2.744,
      "step": 4
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7045454545454546,
      "eval_loss": 0.7323642373085022,
      "eval_runtime": 0.7105,
      "eval_samples_per_second": 61.928,
      "eval_steps_per_second": 2.815,
      "step": 6
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.7272727272727273,
      "eval_loss": 0.5213236808776855,
      "eval_runtime": 0.6833,
      "eval_samples_per_second": 64.389,
      "eval_steps_per_second": 2.927,
      "step": 8
    },
    {
      "epoch": 5.0,
      "grad_norm": 12.558238983154297,
      "learning_rate": 2.5e-05,
      "loss": 0.5701,
      "step": 10
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.47172215580940247,
      "eval_runtime": 0.6672,
      "eval_samples_per_second": 65.951,
      "eval_steps_per_second": 2.998,
      "step": 10
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.7045454545454546,
      "eval_loss": 0.5339322686195374,
      "eval_runtime": 0.66,
      "eval_samples_per_second": 66.668,
      "eval_steps_per_second": 3.03,
      "step": 12
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7272727272727273,
      "eval_loss": 0.49592116475105286,
      "eval_runtime": 0.6766,
      "eval_samples_per_second": 65.028,
      "eval_steps_per_second": 2.956,
      "step": 14
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.4085721969604492,
      "eval_runtime": 0.6823,
      "eval_samples_per_second": 64.484,
      "eval_steps_per_second": 2.931,
      "step": 16
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.4038828909397125,
      "eval_runtime": 0.699,
      "eval_samples_per_second": 62.949,
      "eval_steps_per_second": 2.861,
      "step": 18
    },
    {
      "epoch": 10.0,
      "grad_norm": 6.760091304779053,
      "learning_rate": 5e-05,
      "loss": 0.4248,
      "step": 20
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.41064831614494324,
      "eval_runtime": 0.6804,
      "eval_samples_per_second": 64.67,
      "eval_steps_per_second": 2.94,
      "step": 20
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.41078120470046997,
      "eval_runtime": 0.6789,
      "eval_samples_per_second": 64.811,
      "eval_steps_per_second": 2.946,
      "step": 22
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7727272727272727,
      "eval_loss": 0.4606521725654602,
      "eval_runtime": 0.6783,
      "eval_samples_per_second": 64.871,
      "eval_steps_per_second": 2.949,
      "step": 24
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.7727272727272727,
      "eval_loss": 0.4446246922016144,
      "eval_runtime": 0.6942,
      "eval_samples_per_second": 63.381,
      "eval_steps_per_second": 2.881,
      "step": 26
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.3911612629890442,
      "eval_runtime": 0.6731,
      "eval_samples_per_second": 65.372,
      "eval_steps_per_second": 2.971,
      "step": 28
    },
    {
      "epoch": 15.0,
      "grad_norm": 5.551945209503174,
      "learning_rate": 4.722222222222222e-05,
      "loss": 0.3579,
      "step": 30
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.7727272727272727,
      "eval_loss": 0.5182923078536987,
      "eval_runtime": 0.6921,
      "eval_samples_per_second": 63.573,
      "eval_steps_per_second": 2.89,
      "step": 30
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.2991331219673157,
      "eval_runtime": 0.6842,
      "eval_samples_per_second": 64.311,
      "eval_steps_per_second": 2.923,
      "step": 32
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.35867616534233093,
      "eval_runtime": 0.6814,
      "eval_samples_per_second": 64.576,
      "eval_steps_per_second": 2.935,
      "step": 34
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.31098949909210205,
      "eval_runtime": 0.6742,
      "eval_samples_per_second": 65.262,
      "eval_steps_per_second": 2.966,
      "step": 36
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.30843570828437805,
      "eval_runtime": 0.6849,
      "eval_samples_per_second": 64.239,
      "eval_steps_per_second": 2.92,
      "step": 38
    },
    {
      "epoch": 20.0,
      "grad_norm": 12.44377326965332,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.2838,
      "step": 40
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.30788886547088623,
      "eval_runtime": 0.6749,
      "eval_samples_per_second": 65.195,
      "eval_steps_per_second": 2.963,
      "step": 40
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.30334699153900146,
      "eval_runtime": 0.6841,
      "eval_samples_per_second": 64.317,
      "eval_steps_per_second": 2.924,
      "step": 42
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.31255313754081726,
      "eval_runtime": 0.6824,
      "eval_samples_per_second": 64.482,
      "eval_steps_per_second": 2.931,
      "step": 44
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3171485364437103,
      "eval_runtime": 0.6748,
      "eval_samples_per_second": 65.208,
      "eval_steps_per_second": 2.964,
      "step": 46
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.2689073383808136,
      "eval_runtime": 0.6824,
      "eval_samples_per_second": 64.479,
      "eval_steps_per_second": 2.931,
      "step": 48
    },
    {
      "epoch": 25.0,
      "grad_norm": 15.396455764770508,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.2705,
      "step": 50
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.3174770772457123,
      "eval_runtime": 0.6766,
      "eval_samples_per_second": 65.03,
      "eval_steps_per_second": 2.956,
      "step": 50
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.3463535010814667,
      "eval_runtime": 0.6839,
      "eval_samples_per_second": 64.335,
      "eval_steps_per_second": 2.924,
      "step": 52
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.30922409892082214,
      "eval_runtime": 0.6775,
      "eval_samples_per_second": 64.949,
      "eval_steps_per_second": 2.952,
      "step": 54
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.3178049921989441,
      "eval_runtime": 0.6849,
      "eval_samples_per_second": 64.247,
      "eval_steps_per_second": 2.92,
      "step": 56
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.7954545454545454,
      "eval_loss": 0.4106878340244293,
      "eval_runtime": 0.686,
      "eval_samples_per_second": 64.137,
      "eval_steps_per_second": 2.915,
      "step": 58
    },
    {
      "epoch": 30.0,
      "grad_norm": 3.0605578422546387,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.1887,
      "step": 60
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.41510117053985596,
      "eval_runtime": 0.6795,
      "eval_samples_per_second": 64.754,
      "eval_steps_per_second": 2.943,
      "step": 60
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.7954545454545454,
      "eval_loss": 0.5449530482292175,
      "eval_runtime": 0.6818,
      "eval_samples_per_second": 64.531,
      "eval_steps_per_second": 2.933,
      "step": 62
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.28916433453559875,
      "eval_runtime": 0.6775,
      "eval_samples_per_second": 64.949,
      "eval_steps_per_second": 2.952,
      "step": 64
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.40782248973846436,
      "eval_runtime": 0.6781,
      "eval_samples_per_second": 64.888,
      "eval_steps_per_second": 2.949,
      "step": 66
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.28214308619499207,
      "eval_runtime": 0.6766,
      "eval_samples_per_second": 65.031,
      "eval_steps_per_second": 2.956,
      "step": 68
    },
    {
      "epoch": 35.0,
      "grad_norm": 5.467086315155029,
      "learning_rate": 3.611111111111111e-05,
      "loss": 0.1692,
      "step": 70
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.27076256275177,
      "eval_runtime": 0.6777,
      "eval_samples_per_second": 64.923,
      "eval_steps_per_second": 2.951,
      "step": 70
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.26918765902519226,
      "eval_runtime": 0.6881,
      "eval_samples_per_second": 63.94,
      "eval_steps_per_second": 2.906,
      "step": 72
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.28058433532714844,
      "eval_runtime": 0.709,
      "eval_samples_per_second": 62.057,
      "eval_steps_per_second": 2.821,
      "step": 74
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.46127602458000183,
      "eval_runtime": 0.6984,
      "eval_samples_per_second": 63.003,
      "eval_steps_per_second": 2.864,
      "step": 76
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.28873252868652344,
      "eval_runtime": 0.6977,
      "eval_samples_per_second": 63.06,
      "eval_steps_per_second": 2.866,
      "step": 78
    },
    {
      "epoch": 40.0,
      "grad_norm": 6.031607151031494,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1623,
      "step": 80
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.4045729637145996,
      "eval_runtime": 0.6788,
      "eval_samples_per_second": 64.82,
      "eval_steps_per_second": 2.946,
      "step": 80
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.4541547894477844,
      "eval_runtime": 0.7033,
      "eval_samples_per_second": 62.559,
      "eval_steps_per_second": 2.844,
      "step": 82
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.30104902386665344,
      "eval_runtime": 0.6738,
      "eval_samples_per_second": 65.305,
      "eval_steps_per_second": 2.968,
      "step": 84
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.29540228843688965,
      "eval_runtime": 0.6769,
      "eval_samples_per_second": 65.0,
      "eval_steps_per_second": 2.955,
      "step": 86
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.28378191590309143,
      "eval_runtime": 0.6766,
      "eval_samples_per_second": 65.035,
      "eval_steps_per_second": 2.956,
      "step": 88
    },
    {
      "epoch": 45.0,
      "grad_norm": 3.758543014526367,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.1522,
      "step": 90
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.26745036244392395,
      "eval_runtime": 0.6888,
      "eval_samples_per_second": 63.882,
      "eval_steps_per_second": 2.904,
      "step": 90
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.25173479318618774,
      "eval_runtime": 0.6806,
      "eval_samples_per_second": 64.65,
      "eval_steps_per_second": 2.939,
      "step": 92
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.2687262296676636,
      "eval_runtime": 0.6766,
      "eval_samples_per_second": 65.028,
      "eval_steps_per_second": 2.956,
      "step": 94
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.2550629675388336,
      "eval_runtime": 0.6734,
      "eval_samples_per_second": 65.344,
      "eval_steps_per_second": 2.97,
      "step": 96
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.26609110832214355,
      "eval_runtime": 0.676,
      "eval_samples_per_second": 65.09,
      "eval_steps_per_second": 2.959,
      "step": 98
    },
    {
      "epoch": 50.0,
      "grad_norm": 4.199686050415039,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1379,
      "step": 100
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.3506697714328766,
      "eval_runtime": 0.6748,
      "eval_samples_per_second": 65.201,
      "eval_steps_per_second": 2.964,
      "step": 100
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.26287269592285156,
      "eval_runtime": 0.7109,
      "eval_samples_per_second": 61.891,
      "eval_steps_per_second": 2.813,
      "step": 102
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.2697383463382721,
      "eval_runtime": 0.6753,
      "eval_samples_per_second": 65.16,
      "eval_steps_per_second": 2.962,
      "step": 104
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.30814626812934875,
      "eval_runtime": 0.6792,
      "eval_samples_per_second": 64.781,
      "eval_steps_per_second": 2.945,
      "step": 106
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.3850657641887665,
      "eval_runtime": 0.6958,
      "eval_samples_per_second": 63.232,
      "eval_steps_per_second": 2.874,
      "step": 108
    },
    {
      "epoch": 55.0,
      "grad_norm": 4.418337821960449,
      "learning_rate": 2.5e-05,
      "loss": 0.1283,
      "step": 110
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.31036803126335144,
      "eval_runtime": 0.6753,
      "eval_samples_per_second": 65.156,
      "eval_steps_per_second": 2.962,
      "step": 110
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3623625934123993,
      "eval_runtime": 0.6771,
      "eval_samples_per_second": 64.98,
      "eval_steps_per_second": 2.954,
      "step": 112
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.31991133093833923,
      "eval_runtime": 0.6895,
      "eval_samples_per_second": 63.817,
      "eval_steps_per_second": 2.901,
      "step": 114
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.8181818181818182,
      "eval_loss": 0.4963685870170593,
      "eval_runtime": 0.6808,
      "eval_samples_per_second": 64.63,
      "eval_steps_per_second": 2.938,
      "step": 116
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3355633616447449,
      "eval_runtime": 0.6801,
      "eval_samples_per_second": 64.701,
      "eval_steps_per_second": 2.941,
      "step": 118
    },
    {
      "epoch": 60.0,
      "grad_norm": 4.723308563232422,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1335,
      "step": 120
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.23136349022388458,
      "eval_runtime": 0.6912,
      "eval_samples_per_second": 63.654,
      "eval_steps_per_second": 2.893,
      "step": 120
    },
    {
      "epoch": 61.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.23339642584323883,
      "eval_runtime": 0.6914,
      "eval_samples_per_second": 63.643,
      "eval_steps_per_second": 2.893,
      "step": 122
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.3961027264595032,
      "eval_runtime": 0.6813,
      "eval_samples_per_second": 64.578,
      "eval_steps_per_second": 2.935,
      "step": 124
    },
    {
      "epoch": 63.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.34530192613601685,
      "eval_runtime": 0.7119,
      "eval_samples_per_second": 61.803,
      "eval_steps_per_second": 2.809,
      "step": 126
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.2806227505207062,
      "eval_runtime": 0.6808,
      "eval_samples_per_second": 64.628,
      "eval_steps_per_second": 2.938,
      "step": 128
    },
    {
      "epoch": 65.0,
      "grad_norm": 6.2539143562316895,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.1353,
      "step": 130
    },
    {
      "epoch": 65.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.33719494938850403,
      "eval_runtime": 0.6923,
      "eval_samples_per_second": 63.561,
      "eval_steps_per_second": 2.889,
      "step": 130
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.26754167675971985,
      "eval_runtime": 0.6936,
      "eval_samples_per_second": 63.438,
      "eval_steps_per_second": 2.884,
      "step": 132
    },
    {
      "epoch": 67.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3481806814670563,
      "eval_runtime": 0.6788,
      "eval_samples_per_second": 64.822,
      "eval_steps_per_second": 2.946,
      "step": 134
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.37249478697776794,
      "eval_runtime": 0.6799,
      "eval_samples_per_second": 64.711,
      "eval_steps_per_second": 2.941,
      "step": 136
    },
    {
      "epoch": 69.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.37690678238868713,
      "eval_runtime": 0.6885,
      "eval_samples_per_second": 63.906,
      "eval_steps_per_second": 2.905,
      "step": 138
    },
    {
      "epoch": 70.0,
      "grad_norm": 3.446394920349121,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.099,
      "step": 140
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.8409090909090909,
      "eval_loss": 0.5170295238494873,
      "eval_runtime": 0.6865,
      "eval_samples_per_second": 64.097,
      "eval_steps_per_second": 2.913,
      "step": 140
    },
    {
      "epoch": 71.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.4709642231464386,
      "eval_runtime": 0.6833,
      "eval_samples_per_second": 64.396,
      "eval_steps_per_second": 2.927,
      "step": 142
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.3265904486179352,
      "eval_runtime": 0.686,
      "eval_samples_per_second": 64.139,
      "eval_steps_per_second": 2.915,
      "step": 144
    },
    {
      "epoch": 73.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.3390271067619324,
      "eval_runtime": 0.6764,
      "eval_samples_per_second": 65.048,
      "eval_steps_per_second": 2.957,
      "step": 146
    },
    {
      "epoch": 74.0,
      "eval_accuracy": 0.8636363636363636,
      "eval_loss": 0.3051266372203827,
      "eval_runtime": 0.6768,
      "eval_samples_per_second": 65.012,
      "eval_steps_per_second": 2.955,
      "step": 148
    },
    {
      "epoch": 75.0,
      "grad_norm": 3.681328058242798,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.1179,
      "step": 150
    },
    {
      "epoch": 75.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.3030414879322052,
      "eval_runtime": 0.6884,
      "eval_samples_per_second": 63.914,
      "eval_steps_per_second": 2.905,
      "step": 150
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.3207602798938751,
      "eval_runtime": 0.6882,
      "eval_samples_per_second": 63.931,
      "eval_steps_per_second": 2.906,
      "step": 152
    },
    {
      "epoch": 77.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.29536545276641846,
      "eval_runtime": 0.679,
      "eval_samples_per_second": 64.797,
      "eval_steps_per_second": 2.945,
      "step": 154
    },
    {
      "epoch": 78.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.2776944935321808,
      "eval_runtime": 0.6795,
      "eval_samples_per_second": 64.751,
      "eval_steps_per_second": 2.943,
      "step": 156
    },
    {
      "epoch": 79.0,
      "eval_accuracy": 0.9318181818181818,
      "eval_loss": 0.27630314230918884,
      "eval_runtime": 0.6785,
      "eval_samples_per_second": 64.85,
      "eval_steps_per_second": 2.948,
      "step": 158
    },
    {
      "epoch": 80.0,
      "grad_norm": 4.155190467834473,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.1077,
      "step": 160
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.3058790862560272,
      "eval_runtime": 0.6847,
      "eval_samples_per_second": 64.264,
      "eval_steps_per_second": 2.921,
      "step": 160
    },
    {
      "epoch": 81.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3445180654525757,
      "eval_runtime": 0.6804,
      "eval_samples_per_second": 64.664,
      "eval_steps_per_second": 2.939,
      "step": 162
    },
    {
      "epoch": 82.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.32390668988227844,
      "eval_runtime": 0.6819,
      "eval_samples_per_second": 64.53,
      "eval_steps_per_second": 2.933,
      "step": 164
    },
    {
      "epoch": 83.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.31751757860183716,
      "eval_runtime": 0.6747,
      "eval_samples_per_second": 65.218,
      "eval_steps_per_second": 2.964,
      "step": 166
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.321386456489563,
      "eval_runtime": 0.6763,
      "eval_samples_per_second": 65.059,
      "eval_steps_per_second": 2.957,
      "step": 168
    },
    {
      "epoch": 85.0,
      "grad_norm": 5.664576530456543,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.0907,
      "step": 170
    },
    {
      "epoch": 85.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.33133983612060547,
      "eval_runtime": 0.6779,
      "eval_samples_per_second": 64.91,
      "eval_steps_per_second": 2.95,
      "step": 170
    },
    {
      "epoch": 86.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.34923797845840454,
      "eval_runtime": 0.7083,
      "eval_samples_per_second": 62.116,
      "eval_steps_per_second": 2.823,
      "step": 172
    },
    {
      "epoch": 87.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.3643816113471985,
      "eval_runtime": 0.6853,
      "eval_samples_per_second": 64.202,
      "eval_steps_per_second": 2.918,
      "step": 174
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.36371049284935,
      "eval_runtime": 0.6841,
      "eval_samples_per_second": 64.321,
      "eval_steps_per_second": 2.924,
      "step": 176
    },
    {
      "epoch": 89.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.37499192357063293,
      "eval_runtime": 0.7016,
      "eval_samples_per_second": 62.712,
      "eval_steps_per_second": 2.851,
      "step": 178
    },
    {
      "epoch": 90.0,
      "grad_norm": 2.7906877994537354,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0972,
      "step": 180
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.38448479771614075,
      "eval_runtime": 0.6911,
      "eval_samples_per_second": 63.67,
      "eval_steps_per_second": 2.894,
      "step": 180
    },
    {
      "epoch": 91.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.37490519881248474,
      "eval_runtime": 0.6788,
      "eval_samples_per_second": 64.818,
      "eval_steps_per_second": 2.946,
      "step": 182
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3721046447753906,
      "eval_runtime": 0.6763,
      "eval_samples_per_second": 65.06,
      "eval_steps_per_second": 2.957,
      "step": 184
    },
    {
      "epoch": 93.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.3680332899093628,
      "eval_runtime": 0.6756,
      "eval_samples_per_second": 65.126,
      "eval_steps_per_second": 2.96,
      "step": 186
    },
    {
      "epoch": 94.0,
      "eval_accuracy": 0.8863636363636364,
      "eval_loss": 0.36337849497795105,
      "eval_runtime": 0.6839,
      "eval_samples_per_second": 64.335,
      "eval_steps_per_second": 2.924,
      "step": 188
    },
    {
      "epoch": 95.0,
      "grad_norm": 3.5969669818878174,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.0733,
      "step": 190
    },
    {
      "epoch": 95.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.35654938220977783,
      "eval_runtime": 0.6799,
      "eval_samples_per_second": 64.712,
      "eval_steps_per_second": 2.941,
      "step": 190
    },
    {
      "epoch": 96.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.3518596589565277,
      "eval_runtime": 0.6816,
      "eval_samples_per_second": 64.555,
      "eval_steps_per_second": 2.934,
      "step": 192
    },
    {
      "epoch": 97.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.35288307070732117,
      "eval_runtime": 0.671,
      "eval_samples_per_second": 65.575,
      "eval_steps_per_second": 2.981,
      "step": 194
    },
    {
      "epoch": 98.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.35361969470977783,
      "eval_runtime": 0.6754,
      "eval_samples_per_second": 65.148,
      "eval_steps_per_second": 2.961,
      "step": 196
    },
    {
      "epoch": 99.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.35606831312179565,
      "eval_runtime": 0.675,
      "eval_samples_per_second": 65.187,
      "eval_steps_per_second": 2.963,
      "step": 198
    },
    {
      "epoch": 100.0,
      "grad_norm": 2.4339709281921387,
      "learning_rate": 0.0,
      "loss": 0.079,
      "step": 200
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.9090909090909091,
      "eval_loss": 0.3565208315849304,
      "eval_runtime": 0.6892,
      "eval_samples_per_second": 63.846,
      "eval_steps_per_second": 2.902,
      "step": 200
    },
    {
      "epoch": 100.0,
      "step": 200,
      "total_flos": 1.8822505078960128e+18,
      "train_loss": 0.18895462423563003,
      "train_runtime": 1214.4267,
      "train_samples_per_second": 20.009,
      "train_steps_per_second": 0.165
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.9318181818181818,
      "eval_loss": 0.27630314230918884,
      "eval_runtime": 0.7009,
      "eval_samples_per_second": 62.779,
      "eval_steps_per_second": 2.854,
      "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
}