{
  "best_metric": 0.8142857142857143,
  "best_model_checkpoint": "swin-tiny-patch4-window7-224-finetuned-ADC-3cls-0922/checkpoint-310",
  "epoch": 200.0,
  "eval_steps": 500,
  "global_step": 400,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0693532228469849,
      "eval_runtime": 1.0286,
      "eval_samples_per_second": 68.056,
      "eval_steps_per_second": 1.944,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0689432621002197,
      "eval_runtime": 1.4012,
      "eval_samples_per_second": 49.956,
      "eval_steps_per_second": 1.427,
      "step": 4
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0682058334350586,
      "eval_runtime": 0.8857,
      "eval_samples_per_second": 79.031,
      "eval_steps_per_second": 2.258,
      "step": 6
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0670955181121826,
      "eval_runtime": 1.6997,
      "eval_samples_per_second": 41.183,
      "eval_steps_per_second": 1.177,
      "step": 8
    },
    {
      "epoch": 5.0,
      "learning_rate": 1.25e-05,
      "loss": 1.096,
      "step": 10
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.42857142857142855,
      "eval_loss": 1.065665602684021,
      "eval_runtime": 0.6117,
      "eval_samples_per_second": 114.441,
      "eval_steps_per_second": 3.27,
      "step": 10
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.42857142857142855,
      "eval_loss": 1.0639806985855103,
      "eval_runtime": 0.6447,
      "eval_samples_per_second": 108.583,
      "eval_steps_per_second": 3.102,
      "step": 12
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0620665550231934,
      "eval_runtime": 0.806,
      "eval_samples_per_second": 86.844,
      "eval_steps_per_second": 2.481,
      "step": 14
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.4,
      "eval_loss": 1.0598403215408325,
      "eval_runtime": 0.6374,
      "eval_samples_per_second": 109.817,
      "eval_steps_per_second": 3.138,
      "step": 16
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.4,
      "eval_loss": 1.0572247505187988,
      "eval_runtime": 0.6352,
      "eval_samples_per_second": 110.2,
      "eval_steps_per_second": 3.149,
      "step": 18
    },
    {
      "epoch": 10.0,
      "learning_rate": 2.5e-05,
      "loss": 1.0906,
      "step": 20
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.4,
      "eval_loss": 1.0545086860656738,
      "eval_runtime": 0.8412,
      "eval_samples_per_second": 83.211,
      "eval_steps_per_second": 2.377,
      "step": 20
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0516999959945679,
      "eval_runtime": 0.6382,
      "eval_samples_per_second": 109.685,
      "eval_steps_per_second": 3.134,
      "step": 22
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0486340522766113,
      "eval_runtime": 0.6383,
      "eval_samples_per_second": 109.67,
      "eval_steps_per_second": 3.133,
      "step": 24
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.045298457145691,
      "eval_runtime": 0.8655,
      "eval_samples_per_second": 80.883,
      "eval_steps_per_second": 2.311,
      "step": 26
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0417920351028442,
      "eval_runtime": 0.64,
      "eval_samples_per_second": 109.372,
      "eval_steps_per_second": 3.125,
      "step": 28
    },
    {
      "epoch": 15.0,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 1.0647,
      "step": 30
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0380207300186157,
      "eval_runtime": 0.6461,
      "eval_samples_per_second": 108.335,
      "eval_steps_per_second": 3.095,
      "step": 30
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.0343334674835205,
      "eval_runtime": 0.8283,
      "eval_samples_per_second": 84.515,
      "eval_steps_per_second": 2.415,
      "step": 32
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.4142857142857143,
      "eval_loss": 1.030653953552246,
      "eval_runtime": 0.6353,
      "eval_samples_per_second": 110.181,
      "eval_steps_per_second": 3.148,
      "step": 34
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.42857142857142855,
      "eval_loss": 1.0267900228500366,
      "eval_runtime": 0.6316,
      "eval_samples_per_second": 110.829,
      "eval_steps_per_second": 3.167,
      "step": 36
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.42857142857142855,
      "eval_loss": 1.0229403972625732,
      "eval_runtime": 1.4107,
      "eval_samples_per_second": 49.619,
      "eval_steps_per_second": 1.418,
      "step": 38
    },
    {
      "epoch": 20.0,
      "learning_rate": 5e-05,
      "loss": 1.0451,
      "step": 40
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.44285714285714284,
      "eval_loss": 1.0190969705581665,
      "eval_runtime": 0.6384,
      "eval_samples_per_second": 109.648,
      "eval_steps_per_second": 3.133,
      "step": 40
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.45714285714285713,
      "eval_loss": 1.0152583122253418,
      "eval_runtime": 0.6302,
      "eval_samples_per_second": 111.074,
      "eval_steps_per_second": 3.174,
      "step": 42
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 1.011588454246521,
      "eval_runtime": 0.8016,
      "eval_samples_per_second": 87.325,
      "eval_steps_per_second": 2.495,
      "step": 44
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 1.0081889629364014,
      "eval_runtime": 0.6258,
      "eval_samples_per_second": 111.852,
      "eval_steps_per_second": 3.196,
      "step": 46
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 1.0049412250518799,
      "eval_runtime": 0.6247,
      "eval_samples_per_second": 112.06,
      "eval_steps_per_second": 3.202,
      "step": 48
    },
    {
      "epoch": 25.0,
      "learning_rate": 6.25e-05,
      "loss": 1.037,
      "step": 50
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 1.0015885829925537,
      "eval_runtime": 0.8268,
      "eval_samples_per_second": 84.66,
      "eval_steps_per_second": 2.419,
      "step": 50
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 0.9978756308555603,
      "eval_runtime": 0.635,
      "eval_samples_per_second": 110.241,
      "eval_steps_per_second": 3.15,
      "step": 52
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 0.9943951368331909,
      "eval_runtime": 0.6443,
      "eval_samples_per_second": 108.649,
      "eval_steps_per_second": 3.104,
      "step": 54
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 0.9912726283073425,
      "eval_runtime": 0.8063,
      "eval_samples_per_second": 86.821,
      "eval_steps_per_second": 2.481,
      "step": 56
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 0.9882513880729675,
      "eval_runtime": 0.635,
      "eval_samples_per_second": 110.236,
      "eval_steps_per_second": 3.15,
      "step": 58
    },
    {
      "epoch": 30.0,
      "learning_rate": 7.500000000000001e-05,
      "loss": 1.0214,
      "step": 60
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 0.9846696257591248,
      "eval_runtime": 0.6341,
      "eval_samples_per_second": 110.395,
      "eval_steps_per_second": 3.154,
      "step": 60
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.45714285714285713,
      "eval_loss": 0.9809077382087708,
      "eval_runtime": 0.8112,
      "eval_samples_per_second": 86.297,
      "eval_steps_per_second": 2.466,
      "step": 62
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 0.9768250584602356,
      "eval_runtime": 0.6357,
      "eval_samples_per_second": 110.106,
      "eval_steps_per_second": 3.146,
      "step": 64
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 0.9722528457641602,
      "eval_runtime": 0.6387,
      "eval_samples_per_second": 109.603,
      "eval_steps_per_second": 3.132,
      "step": 66
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 0.9670786261558533,
      "eval_runtime": 0.83,
      "eval_samples_per_second": 84.335,
      "eval_steps_per_second": 2.41,
      "step": 68
    },
    {
      "epoch": 35.0,
      "learning_rate": 8.75e-05,
      "loss": 1.0181,
      "step": 70
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.4714285714285714,
      "eval_loss": 0.9616244435310364,
      "eval_runtime": 0.6343,
      "eval_samples_per_second": 110.364,
      "eval_steps_per_second": 3.153,
      "step": 70
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.4857142857142857,
      "eval_loss": 0.9561253190040588,
      "eval_runtime": 0.6551,
      "eval_samples_per_second": 106.853,
      "eval_steps_per_second": 3.053,
      "step": 72
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.5,
      "eval_loss": 0.950469434261322,
      "eval_runtime": 0.8176,
      "eval_samples_per_second": 85.619,
      "eval_steps_per_second": 2.446,
      "step": 74
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.5285714285714286,
      "eval_loss": 0.9445714950561523,
      "eval_runtime": 0.6352,
      "eval_samples_per_second": 110.194,
      "eval_steps_per_second": 3.148,
      "step": 76
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.5285714285714286,
      "eval_loss": 0.9387871026992798,
      "eval_runtime": 0.629,
      "eval_samples_per_second": 111.283,
      "eval_steps_per_second": 3.18,
      "step": 78
    },
    {
      "epoch": 40.0,
      "learning_rate": 0.0001,
      "loss": 0.9646,
      "step": 80
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.5285714285714286,
      "eval_loss": 0.9331315159797668,
      "eval_runtime": 0.8121,
      "eval_samples_per_second": 86.194,
      "eval_steps_per_second": 2.463,
      "step": 80
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.5142857142857142,
      "eval_loss": 0.9276390075683594,
      "eval_runtime": 0.6356,
      "eval_samples_per_second": 110.129,
      "eval_steps_per_second": 3.147,
      "step": 82
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.5285714285714286,
      "eval_loss": 0.9224144220352173,
      "eval_runtime": 0.6341,
      "eval_samples_per_second": 110.396,
      "eval_steps_per_second": 3.154,
      "step": 84
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.5285714285714286,
      "eval_loss": 0.917235791683197,
      "eval_runtime": 0.8135,
      "eval_samples_per_second": 86.048,
      "eval_steps_per_second": 2.459,
      "step": 86
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.5285714285714286,
      "eval_loss": 0.9120003581047058,
      "eval_runtime": 0.6333,
      "eval_samples_per_second": 110.541,
      "eval_steps_per_second": 3.158,
      "step": 88
    },
    {
      "epoch": 45.0,
      "learning_rate": 9.687500000000001e-05,
      "loss": 0.946,
      "step": 90
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.5142857142857142,
      "eval_loss": 0.9070144891738892,
      "eval_runtime": 0.6382,
      "eval_samples_per_second": 109.685,
      "eval_steps_per_second": 3.134,
      "step": 90
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.5285714285714286,
      "eval_loss": 0.9021272659301758,
      "eval_runtime": 0.8166,
      "eval_samples_per_second": 85.719,
      "eval_steps_per_second": 2.449,
      "step": 92
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.5428571428571428,
      "eval_loss": 0.897597074508667,
      "eval_runtime": 0.634,
      "eval_samples_per_second": 110.411,
      "eval_steps_per_second": 3.155,
      "step": 94
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.5428571428571428,
      "eval_loss": 0.8932848572731018,
      "eval_runtime": 0.6408,
      "eval_samples_per_second": 109.231,
      "eval_steps_per_second": 3.121,
      "step": 96
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 0.8890884518623352,
      "eval_runtime": 0.8162,
      "eval_samples_per_second": 85.76,
      "eval_steps_per_second": 2.45,
      "step": 98
    },
    {
      "epoch": 50.0,
      "learning_rate": 9.375e-05,
      "loss": 0.9244,
      "step": 100
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 0.8845995664596558,
      "eval_runtime": 0.6368,
      "eval_samples_per_second": 109.929,
      "eval_steps_per_second": 3.141,
      "step": 100
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 0.8802858591079712,
      "eval_runtime": 0.6401,
      "eval_samples_per_second": 109.353,
      "eval_steps_per_second": 3.124,
      "step": 102
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 0.8758621215820312,
      "eval_runtime": 0.8118,
      "eval_samples_per_second": 86.227,
      "eval_steps_per_second": 2.464,
      "step": 104
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 0.8715727925300598,
      "eval_runtime": 0.6434,
      "eval_samples_per_second": 108.8,
      "eval_steps_per_second": 3.109,
      "step": 106
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.5714285714285714,
      "eval_loss": 0.8674018383026123,
      "eval_runtime": 0.63,
      "eval_samples_per_second": 111.108,
      "eval_steps_per_second": 3.175,
      "step": 108
    },
    {
      "epoch": 55.0,
      "learning_rate": 9.062500000000001e-05,
      "loss": 0.9228,
      "step": 110
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.5857142857142857,
      "eval_loss": 0.86343914270401,
      "eval_runtime": 0.7324,
      "eval_samples_per_second": 95.574,
      "eval_steps_per_second": 2.731,
      "step": 110
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.6,
      "eval_loss": 0.8597754240036011,
      "eval_runtime": 0.6359,
      "eval_samples_per_second": 110.082,
      "eval_steps_per_second": 3.145,
      "step": 112
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.5857142857142857,
      "eval_loss": 0.8562148213386536,
      "eval_runtime": 0.6308,
      "eval_samples_per_second": 110.969,
      "eval_steps_per_second": 3.171,
      "step": 114
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.6,
      "eval_loss": 0.852704644203186,
      "eval_runtime": 0.6483,
      "eval_samples_per_second": 107.968,
      "eval_steps_per_second": 3.085,
      "step": 116
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.6,
      "eval_loss": 0.8491949439048767,
      "eval_runtime": 0.7037,
      "eval_samples_per_second": 99.48,
      "eval_steps_per_second": 2.842,
      "step": 118
    },
    {
      "epoch": 60.0,
      "learning_rate": 8.75e-05,
      "loss": 0.8956,
      "step": 120
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.6142857142857143,
      "eval_loss": 0.8456201553344727,
      "eval_runtime": 0.6463,
      "eval_samples_per_second": 108.309,
      "eval_steps_per_second": 3.095,
      "step": 120
    },
    {
      "epoch": 61.0,
      "eval_accuracy": 0.6,
      "eval_loss": 0.8420506715774536,
      "eval_runtime": 0.63,
      "eval_samples_per_second": 111.119,
      "eval_steps_per_second": 3.175,
      "step": 122
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.6,
      "eval_loss": 0.8385196924209595,
      "eval_runtime": 0.7958,
      "eval_samples_per_second": 87.963,
      "eval_steps_per_second": 2.513,
      "step": 124
    },
    {
      "epoch": 63.0,
      "eval_accuracy": 0.6,
      "eval_loss": 0.8351073861122131,
      "eval_runtime": 0.6308,
      "eval_samples_per_second": 110.971,
      "eval_steps_per_second": 3.171,
      "step": 126
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.6142857142857143,
      "eval_loss": 0.8317676186561584,
      "eval_runtime": 0.6457,
      "eval_samples_per_second": 108.412,
      "eval_steps_per_second": 3.097,
      "step": 128
    },
    {
      "epoch": 65.0,
      "learning_rate": 8.4375e-05,
      "loss": 0.8943,
      "step": 130
    },
    {
      "epoch": 65.0,
      "eval_accuracy": 0.6142857142857143,
      "eval_loss": 0.8285678029060364,
      "eval_runtime": 0.8132,
      "eval_samples_per_second": 86.08,
      "eval_steps_per_second": 2.459,
      "step": 130
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.6,
      "eval_loss": 0.825462281703949,
      "eval_runtime": 0.6417,
      "eval_samples_per_second": 109.078,
      "eval_steps_per_second": 3.117,
      "step": 132
    },
    {
      "epoch": 67.0,
      "eval_accuracy": 0.6285714285714286,
      "eval_loss": 0.8222988247871399,
      "eval_runtime": 0.6393,
      "eval_samples_per_second": 109.486,
      "eval_steps_per_second": 3.128,
      "step": 134
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.6428571428571429,
      "eval_loss": 0.8190925121307373,
      "eval_runtime": 0.8213,
      "eval_samples_per_second": 85.231,
      "eval_steps_per_second": 2.435,
      "step": 136
    },
    {
      "epoch": 69.0,
      "eval_accuracy": 0.6285714285714286,
      "eval_loss": 0.8158699870109558,
      "eval_runtime": 0.6409,
      "eval_samples_per_second": 109.217,
      "eval_steps_per_second": 3.12,
      "step": 138
    },
    {
      "epoch": 70.0,
      "learning_rate": 8.125000000000001e-05,
      "loss": 0.854,
      "step": 140
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.6428571428571429,
      "eval_loss": 0.8128588199615479,
      "eval_runtime": 0.6403,
      "eval_samples_per_second": 109.332,
      "eval_steps_per_second": 3.124,
      "step": 140
    },
    {
      "epoch": 71.0,
      "eval_accuracy": 0.6714285714285714,
      "eval_loss": 0.8099709749221802,
      "eval_runtime": 0.8076,
      "eval_samples_per_second": 86.675,
      "eval_steps_per_second": 2.476,
      "step": 142
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.6714285714285714,
      "eval_loss": 0.8072643280029297,
      "eval_runtime": 0.6395,
      "eval_samples_per_second": 109.469,
      "eval_steps_per_second": 3.128,
      "step": 144
    },
    {
      "epoch": 73.0,
      "eval_accuracy": 0.6571428571428571,
      "eval_loss": 0.804807722568512,
      "eval_runtime": 0.6346,
      "eval_samples_per_second": 110.311,
      "eval_steps_per_second": 3.152,
      "step": 146
    },
    {
      "epoch": 74.0,
      "eval_accuracy": 0.6714285714285714,
      "eval_loss": 0.8025286793708801,
      "eval_runtime": 0.7964,
      "eval_samples_per_second": 87.9,
      "eval_steps_per_second": 2.511,
      "step": 148
    },
    {
      "epoch": 75.0,
      "learning_rate": 7.8125e-05,
      "loss": 0.8615,
      "step": 150
    },
    {
      "epoch": 75.0,
      "eval_accuracy": 0.6571428571428571,
      "eval_loss": 0.8000553250312805,
      "eval_runtime": 0.6302,
      "eval_samples_per_second": 111.077,
      "eval_steps_per_second": 3.174,
      "step": 150
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.6571428571428571,
      "eval_loss": 0.797595739364624,
      "eval_runtime": 0.6315,
      "eval_samples_per_second": 110.841,
      "eval_steps_per_second": 3.167,
      "step": 152
    },
    {
      "epoch": 77.0,
      "eval_accuracy": 0.6571428571428571,
      "eval_loss": 0.795224130153656,
      "eval_runtime": 0.7988,
      "eval_samples_per_second": 87.627,
      "eval_steps_per_second": 2.504,
      "step": 154
    },
    {
      "epoch": 78.0,
      "eval_accuracy": 0.6571428571428571,
      "eval_loss": 0.7928095459938049,
      "eval_runtime": 0.6405,
      "eval_samples_per_second": 109.291,
      "eval_steps_per_second": 3.123,
      "step": 156
    },
    {
      "epoch": 79.0,
      "eval_accuracy": 0.6571428571428571,
      "eval_loss": 0.7904472351074219,
      "eval_runtime": 0.6375,
      "eval_samples_per_second": 109.802,
      "eval_steps_per_second": 3.137,
      "step": 158
    },
    {
      "epoch": 80.0,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.8507,
      "step": 160
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.6714285714285714,
      "eval_loss": 0.7881516218185425,
      "eval_runtime": 0.8024,
      "eval_samples_per_second": 87.237,
      "eval_steps_per_second": 2.492,
      "step": 160
    },
    {
      "epoch": 81.0,
      "eval_accuracy": 0.6714285714285714,
      "eval_loss": 0.7857871055603027,
      "eval_runtime": 0.6344,
      "eval_samples_per_second": 110.338,
      "eval_steps_per_second": 3.153,
      "step": 162
    },
    {
      "epoch": 82.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7834708094596863,
      "eval_runtime": 0.6376,
      "eval_samples_per_second": 109.781,
      "eval_steps_per_second": 3.137,
      "step": 164
    },
    {
      "epoch": 83.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7811411023139954,
      "eval_runtime": 0.8171,
      "eval_samples_per_second": 85.665,
      "eval_steps_per_second": 2.448,
      "step": 166
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7788016200065613,
      "eval_runtime": 0.6448,
      "eval_samples_per_second": 108.552,
      "eval_steps_per_second": 3.101,
      "step": 168
    },
    {
      "epoch": 85.0,
      "learning_rate": 7.1875e-05,
      "loss": 0.838,
      "step": 170
    },
    {
      "epoch": 85.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7765094041824341,
      "eval_runtime": 0.6436,
      "eval_samples_per_second": 108.77,
      "eval_steps_per_second": 3.108,
      "step": 170
    },
    {
      "epoch": 86.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7743098139762878,
      "eval_runtime": 0.8392,
      "eval_samples_per_second": 83.412,
      "eval_steps_per_second": 2.383,
      "step": 172
    },
    {
      "epoch": 87.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7722915410995483,
      "eval_runtime": 0.6421,
      "eval_samples_per_second": 109.013,
      "eval_steps_per_second": 3.115,
      "step": 174
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7703205943107605,
      "eval_runtime": 0.6583,
      "eval_samples_per_second": 106.337,
      "eval_steps_per_second": 3.038,
      "step": 176
    },
    {
      "epoch": 89.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.768402636051178,
      "eval_runtime": 0.8092,
      "eval_samples_per_second": 86.507,
      "eval_steps_per_second": 2.472,
      "step": 178
    },
    {
      "epoch": 90.0,
      "learning_rate": 6.875e-05,
      "loss": 0.8245,
      "step": 180
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7664420008659363,
      "eval_runtime": 0.6321,
      "eval_samples_per_second": 110.749,
      "eval_steps_per_second": 3.164,
      "step": 180
    },
    {
      "epoch": 91.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7643933296203613,
      "eval_runtime": 0.6416,
      "eval_samples_per_second": 109.102,
      "eval_steps_per_second": 3.117,
      "step": 182
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.6857142857142857,
      "eval_loss": 0.7624573111534119,
      "eval_runtime": 0.8206,
      "eval_samples_per_second": 85.305,
      "eval_steps_per_second": 2.437,
      "step": 184
    },
    {
      "epoch": 93.0,
      "eval_accuracy": 0.7142857142857143,
      "eval_loss": 0.7605774998664856,
      "eval_runtime": 0.6303,
      "eval_samples_per_second": 111.066,
      "eval_steps_per_second": 3.173,
      "step": 186
    },
    {
      "epoch": 94.0,
      "eval_accuracy": 0.7142857142857143,
      "eval_loss": 0.7587181329727173,
      "eval_runtime": 0.6487,
      "eval_samples_per_second": 107.907,
      "eval_steps_per_second": 3.083,
      "step": 188
    },
    {
      "epoch": 95.0,
      "learning_rate": 6.562500000000001e-05,
      "loss": 0.8124,
      "step": 190
    },
    {
      "epoch": 95.0,
      "eval_accuracy": 0.7142857142857143,
      "eval_loss": 0.75688236951828,
      "eval_runtime": 0.83,
      "eval_samples_per_second": 84.337,
      "eval_steps_per_second": 2.41,
      "step": 190
    },
    {
      "epoch": 96.0,
      "eval_accuracy": 0.7285714285714285,
      "eval_loss": 0.755053699016571,
      "eval_runtime": 0.6434,
      "eval_samples_per_second": 108.8,
      "eval_steps_per_second": 3.109,
      "step": 192
    },
    {
      "epoch": 97.0,
      "eval_accuracy": 0.7285714285714285,
      "eval_loss": 0.7533228993415833,
      "eval_runtime": 0.6338,
      "eval_samples_per_second": 110.439,
      "eval_steps_per_second": 3.155,
      "step": 194
    },
    {
      "epoch": 98.0,
      "eval_accuracy": 0.7285714285714285,
      "eval_loss": 0.7516511082649231,
      "eval_runtime": 0.8125,
      "eval_samples_per_second": 86.157,
      "eval_steps_per_second": 2.462,
      "step": 196
    },
    {
      "epoch": 99.0,
      "eval_accuracy": 0.7428571428571429,
      "eval_loss": 0.7499614953994751,
      "eval_runtime": 0.6565,
      "eval_samples_per_second": 106.632,
      "eval_steps_per_second": 3.047,
      "step": 198
    },
    {
      "epoch": 100.0,
      "learning_rate": 6.25e-05,
      "loss": 0.8102,
      "step": 200
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.7428571428571429,
      "eval_loss": 0.7482544183731079,
      "eval_runtime": 0.6344,
      "eval_samples_per_second": 110.346,
      "eval_steps_per_second": 3.153,
      "step": 200
    },
    {
      "epoch": 101.0,
      "eval_accuracy": 0.7428571428571429,
      "eval_loss": 0.7465088963508606,
      "eval_runtime": 0.8175,
      "eval_samples_per_second": 85.623,
      "eval_steps_per_second": 2.446,
      "step": 202
    },
    {
      "epoch": 102.0,
      "eval_accuracy": 0.7428571428571429,
      "eval_loss": 0.7449584007263184,
      "eval_runtime": 0.6372,
      "eval_samples_per_second": 109.854,
      "eval_steps_per_second": 3.139,
      "step": 204
    },
    {
      "epoch": 103.0,
      "eval_accuracy": 0.7428571428571429,
      "eval_loss": 0.7434430718421936,
      "eval_runtime": 0.6379,
      "eval_samples_per_second": 109.741,
      "eval_steps_per_second": 3.135,
      "step": 206
    },
    {
      "epoch": 104.0,
      "eval_accuracy": 0.7428571428571429,
      "eval_loss": 0.7419188618659973,
      "eval_runtime": 0.8331,
      "eval_samples_per_second": 84.027,
      "eval_steps_per_second": 2.401,
      "step": 208
    },
    {
      "epoch": 105.0,
      "learning_rate": 5.9375e-05,
      "loss": 0.821,
      "step": 210
    },
    {
      "epoch": 105.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7403832674026489,
      "eval_runtime": 0.6317,
      "eval_samples_per_second": 110.817,
      "eval_steps_per_second": 3.166,
      "step": 210
    },
    {
      "epoch": 106.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7388736605644226,
      "eval_runtime": 0.638,
      "eval_samples_per_second": 109.722,
      "eval_steps_per_second": 3.135,
      "step": 212
    },
    {
      "epoch": 107.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7374056577682495,
      "eval_runtime": 0.8271,
      "eval_samples_per_second": 84.63,
      "eval_steps_per_second": 2.418,
      "step": 214
    },
    {
      "epoch": 108.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7359411120414734,
      "eval_runtime": 0.6382,
      "eval_samples_per_second": 109.678,
      "eval_steps_per_second": 3.134,
      "step": 216
    },
    {
      "epoch": 109.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.734478235244751,
      "eval_runtime": 0.6424,
      "eval_samples_per_second": 108.966,
      "eval_steps_per_second": 3.113,
      "step": 218
    },
    {
      "epoch": 110.0,
      "learning_rate": 5.6250000000000005e-05,
      "loss": 0.7918,
      "step": 220
    },
    {
      "epoch": 110.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7330225110054016,
      "eval_runtime": 0.8294,
      "eval_samples_per_second": 84.398,
      "eval_steps_per_second": 2.411,
      "step": 220
    },
    {
      "epoch": 111.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7315928339958191,
      "eval_runtime": 0.6456,
      "eval_samples_per_second": 108.432,
      "eval_steps_per_second": 3.098,
      "step": 222
    },
    {
      "epoch": 112.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7302229404449463,
      "eval_runtime": 0.6368,
      "eval_samples_per_second": 109.931,
      "eval_steps_per_second": 3.141,
      "step": 224
    },
    {
      "epoch": 113.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7288532257080078,
      "eval_runtime": 0.8298,
      "eval_samples_per_second": 84.362,
      "eval_steps_per_second": 2.41,
      "step": 226
    },
    {
      "epoch": 114.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7275059223175049,
      "eval_runtime": 0.6447,
      "eval_samples_per_second": 108.584,
      "eval_steps_per_second": 3.102,
      "step": 228
    },
    {
      "epoch": 115.0,
      "learning_rate": 5.3125000000000004e-05,
      "loss": 0.8063,
      "step": 230
    },
    {
      "epoch": 115.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.7261765599250793,
      "eval_runtime": 0.6376,
      "eval_samples_per_second": 109.779,
      "eval_steps_per_second": 3.137,
      "step": 230
    },
    {
      "epoch": 116.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.7246890068054199,
      "eval_runtime": 0.8093,
      "eval_samples_per_second": 86.498,
      "eval_steps_per_second": 2.471,
      "step": 232
    },
    {
      "epoch": 117.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7232338190078735,
      "eval_runtime": 0.6473,
      "eval_samples_per_second": 108.146,
      "eval_steps_per_second": 3.09,
      "step": 234
    },
    {
      "epoch": 118.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7218143939971924,
      "eval_runtime": 0.639,
      "eval_samples_per_second": 109.553,
      "eval_steps_per_second": 3.13,
      "step": 236
    },
    {
      "epoch": 119.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7204232811927795,
      "eval_runtime": 0.8245,
      "eval_samples_per_second": 84.902,
      "eval_steps_per_second": 2.426,
      "step": 238
    },
    {
      "epoch": 120.0,
      "learning_rate": 5e-05,
      "loss": 0.7897,
      "step": 240
    },
    {
      "epoch": 120.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7191569209098816,
      "eval_runtime": 0.639,
      "eval_samples_per_second": 109.548,
      "eval_steps_per_second": 3.13,
      "step": 240
    },
    {
      "epoch": 121.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7179904580116272,
      "eval_runtime": 0.6313,
      "eval_samples_per_second": 110.887,
      "eval_steps_per_second": 3.168,
      "step": 242
    },
    {
      "epoch": 122.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7168429493904114,
      "eval_runtime": 0.8342,
      "eval_samples_per_second": 83.908,
      "eval_steps_per_second": 2.397,
      "step": 244
    },
    {
      "epoch": 123.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7157979011535645,
      "eval_runtime": 0.6333,
      "eval_samples_per_second": 110.541,
      "eval_steps_per_second": 3.158,
      "step": 246
    },
    {
      "epoch": 124.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.714880645275116,
      "eval_runtime": 0.6368,
      "eval_samples_per_second": 109.924,
      "eval_steps_per_second": 3.141,
      "step": 248
    },
    {
      "epoch": 125.0,
      "learning_rate": 4.6875e-05,
      "loss": 0.7845,
      "step": 250
    },
    {
      "epoch": 125.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7140344381332397,
      "eval_runtime": 0.8287,
      "eval_samples_per_second": 84.466,
      "eval_steps_per_second": 2.413,
      "step": 250
    },
    {
      "epoch": 126.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7130730152130127,
      "eval_runtime": 0.6617,
      "eval_samples_per_second": 105.79,
      "eval_steps_per_second": 3.023,
      "step": 252
    },
    {
      "epoch": 127.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7120916247367859,
      "eval_runtime": 0.6392,
      "eval_samples_per_second": 109.509,
      "eval_steps_per_second": 3.129,
      "step": 254
    },
    {
      "epoch": 128.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7110173106193542,
      "eval_runtime": 0.8272,
      "eval_samples_per_second": 84.623,
      "eval_steps_per_second": 2.418,
      "step": 256
    },
    {
      "epoch": 129.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7099365592002869,
      "eval_runtime": 0.8513,
      "eval_samples_per_second": 82.229,
      "eval_steps_per_second": 2.349,
      "step": 258
    },
    {
      "epoch": 130.0,
      "learning_rate": 4.375e-05,
      "loss": 0.7781,
      "step": 260
    },
    {
      "epoch": 130.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7087655663490295,
      "eval_runtime": 0.6529,
      "eval_samples_per_second": 107.222,
      "eval_steps_per_second": 3.063,
      "step": 260
    },
    {
      "epoch": 131.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7076297402381897,
      "eval_runtime": 0.633,
      "eval_samples_per_second": 110.582,
      "eval_steps_per_second": 3.159,
      "step": 262
    },
    {
      "epoch": 132.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.706558883190155,
      "eval_runtime": 0.8161,
      "eval_samples_per_second": 85.772,
      "eval_steps_per_second": 2.451,
      "step": 264
    },
    {
      "epoch": 133.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7055317759513855,
      "eval_runtime": 0.6432,
      "eval_samples_per_second": 108.834,
      "eval_steps_per_second": 3.11,
      "step": 266
    },
    {
      "epoch": 134.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.7044604420661926,
      "eval_runtime": 0.6426,
      "eval_samples_per_second": 108.936,
      "eval_steps_per_second": 3.112,
      "step": 268
    },
    {
      "epoch": 135.0,
      "learning_rate": 4.0625000000000005e-05,
      "loss": 0.7708,
      "step": 270
    },
    {
      "epoch": 135.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.7034193873405457,
      "eval_runtime": 0.8178,
      "eval_samples_per_second": 85.591,
      "eval_steps_per_second": 2.445,
      "step": 270
    },
    {
      "epoch": 136.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7024958729743958,
      "eval_runtime": 0.6353,
      "eval_samples_per_second": 110.186,
      "eval_steps_per_second": 3.148,
      "step": 272
    },
    {
      "epoch": 137.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7016207575798035,
      "eval_runtime": 0.632,
      "eval_samples_per_second": 110.758,
      "eval_steps_per_second": 3.165,
      "step": 274
    },
    {
      "epoch": 138.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.7007526755332947,
      "eval_runtime": 0.8281,
      "eval_samples_per_second": 84.534,
      "eval_steps_per_second": 2.415,
      "step": 276
    },
    {
      "epoch": 139.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.6998603343963623,
      "eval_runtime": 0.6379,
      "eval_samples_per_second": 109.736,
      "eval_steps_per_second": 3.135,
      "step": 278
    },
    {
      "epoch": 140.0,
      "learning_rate": 3.7500000000000003e-05,
      "loss": 0.797,
      "step": 280
    },
    {
      "epoch": 140.0,
      "eval_accuracy": 0.7571428571428571,
      "eval_loss": 0.6989655494689941,
      "eval_runtime": 0.6404,
      "eval_samples_per_second": 109.303,
      "eval_steps_per_second": 3.123,
      "step": 280
    },
    {
      "epoch": 141.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.6981316804885864,
      "eval_runtime": 0.8143,
      "eval_samples_per_second": 85.963,
      "eval_steps_per_second": 2.456,
      "step": 282
    },
    {
      "epoch": 142.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.6973427534103394,
      "eval_runtime": 0.6382,
      "eval_samples_per_second": 109.681,
      "eval_steps_per_second": 3.134,
      "step": 284
    },
    {
      "epoch": 143.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.6966080069541931,
      "eval_runtime": 0.6394,
      "eval_samples_per_second": 109.471,
      "eval_steps_per_second": 3.128,
      "step": 286
    },
    {
      "epoch": 144.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.6958935856819153,
      "eval_runtime": 0.8183,
      "eval_samples_per_second": 85.546,
      "eval_steps_per_second": 2.444,
      "step": 288
    },
    {
      "epoch": 145.0,
      "learning_rate": 3.4375e-05,
      "loss": 0.7768,
      "step": 290
    },
    {
      "epoch": 145.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.695166289806366,
      "eval_runtime": 0.6458,
      "eval_samples_per_second": 108.386,
      "eval_steps_per_second": 3.097,
      "step": 290
    },
    {
      "epoch": 146.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.694381833076477,
      "eval_runtime": 0.6351,
      "eval_samples_per_second": 110.21,
      "eval_steps_per_second": 3.149,
      "step": 292
    },
    {
      "epoch": 147.0,
      "eval_accuracy": 0.7714285714285715,
      "eval_loss": 0.6935797333717346,
      "eval_runtime": 0.8207,
      "eval_samples_per_second": 85.29,
      "eval_steps_per_second": 2.437,
      "step": 294
    },
    {
      "epoch": 148.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6927558779716492,
      "eval_runtime": 0.6395,
      "eval_samples_per_second": 109.463,
      "eval_steps_per_second": 3.128,
      "step": 296
    },
    {
      "epoch": 149.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6919543147087097,
      "eval_runtime": 0.6322,
      "eval_samples_per_second": 110.727,
      "eval_steps_per_second": 3.164,
      "step": 298
    },
    {
      "epoch": 150.0,
      "learning_rate": 3.125e-05,
      "loss": 0.7569,
      "step": 300
    },
    {
      "epoch": 150.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6911686658859253,
      "eval_runtime": 0.8112,
      "eval_samples_per_second": 86.294,
      "eval_steps_per_second": 2.466,
      "step": 300
    },
    {
      "epoch": 151.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.6904271841049194,
      "eval_runtime": 0.6485,
      "eval_samples_per_second": 107.942,
      "eval_steps_per_second": 3.084,
      "step": 302
    },
    {
      "epoch": 152.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.6896898150444031,
      "eval_runtime": 0.6429,
      "eval_samples_per_second": 108.878,
      "eval_steps_per_second": 3.111,
      "step": 304
    },
    {
      "epoch": 153.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.688970148563385,
      "eval_runtime": 0.8248,
      "eval_samples_per_second": 84.872,
      "eval_steps_per_second": 2.425,
      "step": 306
    },
    {
      "epoch": 154.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.6882473826408386,
      "eval_runtime": 0.6473,
      "eval_samples_per_second": 108.141,
      "eval_steps_per_second": 3.09,
      "step": 308
    },
    {
      "epoch": 155.0,
      "learning_rate": 2.8125000000000003e-05,
      "loss": 0.7807,
      "step": 310
    },
    {
      "epoch": 155.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.6875176429748535,
      "eval_runtime": 0.6211,
      "eval_samples_per_second": 112.707,
      "eval_steps_per_second": 3.22,
      "step": 310
    },
    {
      "epoch": 156.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.6867862939834595,
      "eval_runtime": 0.8104,
      "eval_samples_per_second": 86.38,
      "eval_steps_per_second": 2.468,
      "step": 312
    },
    {
      "epoch": 157.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.6860566139221191,
      "eval_runtime": 0.6376,
      "eval_samples_per_second": 109.79,
      "eval_steps_per_second": 3.137,
      "step": 314
    },
    {
      "epoch": 158.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.6853832006454468,
      "eval_runtime": 0.6414,
      "eval_samples_per_second": 109.144,
      "eval_steps_per_second": 3.118,
      "step": 316
    },
    {
      "epoch": 159.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.684758186340332,
      "eval_runtime": 0.8168,
      "eval_samples_per_second": 85.701,
      "eval_steps_per_second": 2.449,
      "step": 318
    },
    {
      "epoch": 160.0,
      "learning_rate": 2.5e-05,
      "loss": 0.7472,
      "step": 320
    },
    {
      "epoch": 160.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.6841580867767334,
      "eval_runtime": 0.6398,
      "eval_samples_per_second": 109.413,
      "eval_steps_per_second": 3.126,
      "step": 320
    },
    {
      "epoch": 161.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.6836223006248474,
      "eval_runtime": 0.6529,
      "eval_samples_per_second": 107.211,
      "eval_steps_per_second": 3.063,
      "step": 322
    },
    {
      "epoch": 162.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.683104932308197,
      "eval_runtime": 0.8258,
      "eval_samples_per_second": 84.764,
      "eval_steps_per_second": 2.422,
      "step": 324
    },
    {
      "epoch": 163.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.6826251745223999,
      "eval_runtime": 0.6361,
      "eval_samples_per_second": 110.054,
      "eval_steps_per_second": 3.144,
      "step": 326
    },
    {
      "epoch": 164.0,
      "eval_accuracy": 0.8142857142857143,
      "eval_loss": 0.6821797490119934,
      "eval_runtime": 0.6381,
      "eval_samples_per_second": 109.703,
      "eval_steps_per_second": 3.134,
      "step": 328
    },
    {
      "epoch": 165.0,
      "learning_rate": 2.1875e-05,
      "loss": 0.7665,
      "step": 330
    },
    {
      "epoch": 165.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.6817546486854553,
      "eval_runtime": 0.8246,
      "eval_samples_per_second": 84.889,
      "eval_steps_per_second": 2.425,
      "step": 330
    },
    {
      "epoch": 166.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.6813837885856628,
      "eval_runtime": 0.6625,
      "eval_samples_per_second": 105.654,
      "eval_steps_per_second": 3.019,
      "step": 332
    },
    {
      "epoch": 167.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.681039035320282,
      "eval_runtime": 0.6385,
      "eval_samples_per_second": 109.638,
      "eval_steps_per_second": 3.133,
      "step": 334
    },
    {
      "epoch": 168.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6806796193122864,
      "eval_runtime": 0.8294,
      "eval_samples_per_second": 84.403,
      "eval_steps_per_second": 2.412,
      "step": 336
    },
    {
      "epoch": 169.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.680313766002655,
      "eval_runtime": 0.6312,
      "eval_samples_per_second": 110.896,
      "eval_steps_per_second": 3.168,
      "step": 338
    },
    {
      "epoch": 170.0,
      "learning_rate": 1.8750000000000002e-05,
      "loss": 0.7684,
      "step": 340
    },
    {
      "epoch": 170.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6799898743629456,
      "eval_runtime": 0.6322,
      "eval_samples_per_second": 110.721,
      "eval_steps_per_second": 3.163,
      "step": 340
    },
    {
      "epoch": 171.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6796825528144836,
      "eval_runtime": 0.8381,
      "eval_samples_per_second": 83.525,
      "eval_steps_per_second": 2.386,
      "step": 342
    },
    {
      "epoch": 172.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6793543696403503,
      "eval_runtime": 0.6493,
      "eval_samples_per_second": 107.81,
      "eval_steps_per_second": 3.08,
      "step": 344
    },
    {
      "epoch": 173.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6790363192558289,
      "eval_runtime": 0.6451,
      "eval_samples_per_second": 108.504,
      "eval_steps_per_second": 3.1,
      "step": 346
    },
    {
      "epoch": 174.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6787369847297668,
      "eval_runtime": 0.8298,
      "eval_samples_per_second": 84.361,
      "eval_steps_per_second": 2.41,
      "step": 348
    },
    {
      "epoch": 175.0,
      "learning_rate": 1.5625e-05,
      "loss": 0.7459,
      "step": 350
    },
    {
      "epoch": 175.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6784414052963257,
      "eval_runtime": 0.6607,
      "eval_samples_per_second": 105.953,
      "eval_steps_per_second": 3.027,
      "step": 350
    },
    {
      "epoch": 176.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6781107783317566,
      "eval_runtime": 0.681,
      "eval_samples_per_second": 102.789,
      "eval_steps_per_second": 2.937,
      "step": 352
    },
    {
      "epoch": 177.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.677795946598053,
      "eval_runtime": 0.8379,
      "eval_samples_per_second": 83.54,
      "eval_steps_per_second": 2.387,
      "step": 354
    },
    {
      "epoch": 178.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6774783730506897,
      "eval_runtime": 0.6408,
      "eval_samples_per_second": 109.247,
      "eval_steps_per_second": 3.121,
      "step": 356
    },
    {
      "epoch": 179.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6771765351295471,
      "eval_runtime": 0.6403,
      "eval_samples_per_second": 109.315,
      "eval_steps_per_second": 3.123,
      "step": 358
    },
    {
      "epoch": 180.0,
      "learning_rate": 1.25e-05,
      "loss": 0.742,
      "step": 360
    },
    {
      "epoch": 180.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6768958568572998,
      "eval_runtime": 0.8321,
      "eval_samples_per_second": 84.126,
      "eval_steps_per_second": 2.404,
      "step": 360
    },
    {
      "epoch": 181.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6766448616981506,
      "eval_runtime": 0.6364,
      "eval_samples_per_second": 110.0,
      "eval_steps_per_second": 3.143,
      "step": 362
    },
    {
      "epoch": 182.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6764284372329712,
      "eval_runtime": 0.647,
      "eval_samples_per_second": 108.196,
      "eval_steps_per_second": 3.091,
      "step": 364
    },
    {
      "epoch": 183.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6762242913246155,
      "eval_runtime": 0.8597,
      "eval_samples_per_second": 81.422,
      "eval_steps_per_second": 2.326,
      "step": 366
    },
    {
      "epoch": 184.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6760057806968689,
      "eval_runtime": 0.651,
      "eval_samples_per_second": 107.527,
      "eval_steps_per_second": 3.072,
      "step": 368
    },
    {
      "epoch": 185.0,
      "learning_rate": 9.375000000000001e-06,
      "loss": 0.7642,
      "step": 370
    },
    {
      "epoch": 185.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6757904291152954,
      "eval_runtime": 0.6445,
      "eval_samples_per_second": 108.618,
      "eval_steps_per_second": 3.103,
      "step": 370
    },
    {
      "epoch": 186.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6755796670913696,
      "eval_runtime": 0.8335,
      "eval_samples_per_second": 83.986,
      "eval_steps_per_second": 2.4,
      "step": 372
    },
    {
      "epoch": 187.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6753801107406616,
      "eval_runtime": 0.6394,
      "eval_samples_per_second": 109.479,
      "eval_steps_per_second": 3.128,
      "step": 374
    },
    {
      "epoch": 188.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6752031445503235,
      "eval_runtime": 0.658,
      "eval_samples_per_second": 106.386,
      "eval_steps_per_second": 3.04,
      "step": 376
    },
    {
      "epoch": 189.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6750344634056091,
      "eval_runtime": 0.8422,
      "eval_samples_per_second": 83.112,
      "eval_steps_per_second": 2.375,
      "step": 378
    },
    {
      "epoch": 190.0,
      "learning_rate": 6.25e-06,
      "loss": 0.7277,
      "step": 380
    },
    {
      "epoch": 190.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.674885094165802,
      "eval_runtime": 0.6493,
      "eval_samples_per_second": 107.808,
      "eval_steps_per_second": 3.08,
      "step": 380
    },
    {
      "epoch": 191.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6747546195983887,
      "eval_runtime": 0.6429,
      "eval_samples_per_second": 108.875,
      "eval_steps_per_second": 3.111,
      "step": 382
    },
    {
      "epoch": 192.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6746455430984497,
      "eval_runtime": 0.8305,
      "eval_samples_per_second": 84.289,
      "eval_steps_per_second": 2.408,
      "step": 384
    },
    {
      "epoch": 193.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6745493412017822,
      "eval_runtime": 0.6551,
      "eval_samples_per_second": 106.855,
      "eval_steps_per_second": 3.053,
      "step": 386
    },
    {
      "epoch": 194.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6744527816772461,
      "eval_runtime": 0.6428,
      "eval_samples_per_second": 108.896,
      "eval_steps_per_second": 3.111,
      "step": 388
    },
    {
      "epoch": 195.0,
      "learning_rate": 3.125e-06,
      "loss": 0.764,
      "step": 390
    },
    {
      "epoch": 195.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6743654608726501,
      "eval_runtime": 0.8334,
      "eval_samples_per_second": 83.991,
      "eval_steps_per_second": 2.4,
      "step": 390
    },
    {
      "epoch": 196.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.674295961856842,
      "eval_runtime": 0.652,
      "eval_samples_per_second": 107.363,
      "eval_steps_per_second": 3.068,
      "step": 392
    },
    {
      "epoch": 197.0,
      "eval_accuracy": 0.7857142857142857,
      "eval_loss": 0.6742398142814636,
      "eval_runtime": 0.6659,
      "eval_samples_per_second": 105.127,
      "eval_steps_per_second": 3.004,
      "step": 394
    },
    {
      "epoch": 198.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.6741960644721985,
      "eval_runtime": 0.8894,
      "eval_samples_per_second": 78.703,
      "eval_steps_per_second": 2.249,
      "step": 396
    },
    {
      "epoch": 199.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.674168586730957,
      "eval_runtime": 0.6565,
      "eval_samples_per_second": 106.623,
      "eval_steps_per_second": 3.046,
      "step": 398
    },
    {
      "epoch": 200.0,
      "learning_rate": 0.0,
      "loss": 0.7444,
      "step": 400
    },
    {
      "epoch": 200.0,
      "eval_accuracy": 0.8,
      "eval_loss": 0.6741567850112915,
      "eval_runtime": 0.6417,
      "eval_samples_per_second": 109.09,
      "eval_steps_per_second": 3.117,
      "step": 400
    },
    {
      "epoch": 200.0,
      "step": 400,
      "total_flos": 2.23710151698432e+18,
      "train_loss": 0.8548950719833374,
      "train_runtime": 1030.1946,
      "train_samples_per_second": 87.362,
      "train_steps_per_second": 0.388
    }
  ],
  "logging_steps": 10,
  "max_steps": 400,
  "num_train_epochs": 200,
  "save_steps": 500,
  "total_flos": 2.23710151698432e+18,
  "trial_name": null,
  "trial_params": null
}