{
  "best_metric": 0.9534883720930233,
  "best_model_checkpoint": "beit-base-patch16-224-75-fold5/checkpoint-78",
  "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": 0.5116279069767442,
      "eval_loss": 0.6862077116966248,
      "eval_runtime": 0.7154,
      "eval_samples_per_second": 60.105,
      "eval_steps_per_second": 2.796,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.5912647843360901,
      "eval_runtime": 0.6811,
      "eval_samples_per_second": 63.137,
      "eval_steps_per_second": 2.937,
      "step": 4
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.7203647494316101,
      "eval_runtime": 0.6611,
      "eval_samples_per_second": 65.039,
      "eval_steps_per_second": 3.025,
      "step": 6
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.5994660258293152,
      "eval_runtime": 0.6653,
      "eval_samples_per_second": 64.633,
      "eval_steps_per_second": 3.006,
      "step": 8
    },
    {
      "epoch": 5.0,
      "grad_norm": 4.852715969085693,
      "learning_rate": 2.5e-05,
      "loss": 0.6162,
      "step": 10
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.4234636127948761,
      "eval_runtime": 0.6369,
      "eval_samples_per_second": 67.511,
      "eval_steps_per_second": 3.14,
      "step": 10
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.3975236713886261,
      "eval_runtime": 0.6571,
      "eval_samples_per_second": 65.443,
      "eval_steps_per_second": 3.044,
      "step": 12
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7674418604651163,
      "eval_loss": 0.6029347777366638,
      "eval_runtime": 0.6674,
      "eval_samples_per_second": 64.429,
      "eval_steps_per_second": 2.997,
      "step": 14
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.46703392267227173,
      "eval_runtime": 0.6777,
      "eval_samples_per_second": 63.453,
      "eval_steps_per_second": 2.951,
      "step": 16
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.34478357434272766,
      "eval_runtime": 0.6694,
      "eval_samples_per_second": 64.232,
      "eval_steps_per_second": 2.988,
      "step": 18
    },
    {
      "epoch": 10.0,
      "grad_norm": 7.711123466491699,
      "learning_rate": 5e-05,
      "loss": 0.4312,
      "step": 20
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.4463817775249481,
      "eval_runtime": 0.6613,
      "eval_samples_per_second": 65.021,
      "eval_steps_per_second": 3.024,
      "step": 20
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.33956387639045715,
      "eval_runtime": 0.6693,
      "eval_samples_per_second": 64.249,
      "eval_steps_per_second": 2.988,
      "step": 22
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.4007296562194824,
      "eval_runtime": 0.6584,
      "eval_samples_per_second": 65.312,
      "eval_steps_per_second": 3.038,
      "step": 24
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.3398185670375824,
      "eval_runtime": 0.6701,
      "eval_samples_per_second": 64.17,
      "eval_steps_per_second": 2.985,
      "step": 26
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.4276241064071655,
      "eval_runtime": 0.6779,
      "eval_samples_per_second": 63.436,
      "eval_steps_per_second": 2.95,
      "step": 28
    },
    {
      "epoch": 15.0,
      "grad_norm": 4.13345193862915,
      "learning_rate": 4.722222222222222e-05,
      "loss": 0.3453,
      "step": 30
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.43359655141830444,
      "eval_runtime": 0.6629,
      "eval_samples_per_second": 64.864,
      "eval_steps_per_second": 3.017,
      "step": 30
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.3777061402797699,
      "eval_runtime": 0.6673,
      "eval_samples_per_second": 64.442,
      "eval_steps_per_second": 2.997,
      "step": 32
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5909541249275208,
      "eval_runtime": 0.6717,
      "eval_samples_per_second": 64.017,
      "eval_steps_per_second": 2.978,
      "step": 34
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.6095340251922607,
      "eval_runtime": 0.6665,
      "eval_samples_per_second": 64.512,
      "eval_steps_per_second": 3.001,
      "step": 36
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.35701826214790344,
      "eval_runtime": 0.6681,
      "eval_samples_per_second": 64.359,
      "eval_steps_per_second": 2.993,
      "step": 38
    },
    {
      "epoch": 20.0,
      "grad_norm": 6.075029373168945,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.3288,
      "step": 40
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5201966762542725,
      "eval_runtime": 0.6605,
      "eval_samples_per_second": 65.1,
      "eval_steps_per_second": 3.028,
      "step": 40
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5604430437088013,
      "eval_runtime": 0.6623,
      "eval_samples_per_second": 64.929,
      "eval_steps_per_second": 3.02,
      "step": 42
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.2948876619338989,
      "eval_runtime": 0.6627,
      "eval_samples_per_second": 64.885,
      "eval_steps_per_second": 3.018,
      "step": 44
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.3442337214946747,
      "eval_runtime": 0.674,
      "eval_samples_per_second": 63.797,
      "eval_steps_per_second": 2.967,
      "step": 46
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.28198882937431335,
      "eval_runtime": 0.6623,
      "eval_samples_per_second": 64.928,
      "eval_steps_per_second": 3.02,
      "step": 48
    },
    {
      "epoch": 25.0,
      "grad_norm": 3.311546564102173,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.2571,
      "step": 50
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.3239896297454834,
      "eval_runtime": 0.6644,
      "eval_samples_per_second": 64.724,
      "eval_steps_per_second": 3.01,
      "step": 50
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.2908729016780853,
      "eval_runtime": 0.6946,
      "eval_samples_per_second": 61.905,
      "eval_steps_per_second": 2.879,
      "step": 52
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.24289709329605103,
      "eval_runtime": 0.6623,
      "eval_samples_per_second": 64.929,
      "eval_steps_per_second": 3.02,
      "step": 54
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.2280385047197342,
      "eval_runtime": 0.6791,
      "eval_samples_per_second": 63.317,
      "eval_steps_per_second": 2.945,
      "step": 56
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.39840760827064514,
      "eval_runtime": 0.7239,
      "eval_samples_per_second": 59.399,
      "eval_steps_per_second": 2.763,
      "step": 58
    },
    {
      "epoch": 30.0,
      "grad_norm": 7.222219944000244,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.2012,
      "step": 60
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.29045242071151733,
      "eval_runtime": 0.6714,
      "eval_samples_per_second": 64.046,
      "eval_steps_per_second": 2.979,
      "step": 60
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.2508714497089386,
      "eval_runtime": 0.6943,
      "eval_samples_per_second": 61.93,
      "eval_steps_per_second": 2.88,
      "step": 62
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.28884419798851013,
      "eval_runtime": 0.6903,
      "eval_samples_per_second": 62.294,
      "eval_steps_per_second": 2.897,
      "step": 64
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.2689167857170105,
      "eval_runtime": 0.6689,
      "eval_samples_per_second": 64.28,
      "eval_steps_per_second": 2.99,
      "step": 66
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.24166707694530487,
      "eval_runtime": 0.6524,
      "eval_samples_per_second": 65.908,
      "eval_steps_per_second": 3.065,
      "step": 68
    },
    {
      "epoch": 35.0,
      "grad_norm": 6.977738857269287,
      "learning_rate": 3.611111111111111e-05,
      "loss": 0.1814,
      "step": 70
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.24178560078144073,
      "eval_runtime": 0.6548,
      "eval_samples_per_second": 65.669,
      "eval_steps_per_second": 3.054,
      "step": 70
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.24908961355686188,
      "eval_runtime": 0.6723,
      "eval_samples_per_second": 63.959,
      "eval_steps_per_second": 2.975,
      "step": 72
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.2998228669166565,
      "eval_runtime": 0.6685,
      "eval_samples_per_second": 64.321,
      "eval_steps_per_second": 2.992,
      "step": 74
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.2743920683860779,
      "eval_runtime": 0.6739,
      "eval_samples_per_second": 63.812,
      "eval_steps_per_second": 2.968,
      "step": 76
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.9534883720930233,
      "eval_loss": 0.2663915455341339,
      "eval_runtime": 0.6946,
      "eval_samples_per_second": 61.91,
      "eval_steps_per_second": 2.88,
      "step": 78
    },
    {
      "epoch": 40.0,
      "grad_norm": 6.1886067390441895,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1555,
      "step": 80
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.2159864455461502,
      "eval_runtime": 0.6941,
      "eval_samples_per_second": 61.947,
      "eval_steps_per_second": 2.881,
      "step": 80
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.38748297095298767,
      "eval_runtime": 0.6699,
      "eval_samples_per_second": 64.193,
      "eval_steps_per_second": 2.986,
      "step": 82
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.4607594609260559,
      "eval_runtime": 0.697,
      "eval_samples_per_second": 61.689,
      "eval_steps_per_second": 2.869,
      "step": 84
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.2978023886680603,
      "eval_runtime": 0.6715,
      "eval_samples_per_second": 64.036,
      "eval_steps_per_second": 2.978,
      "step": 86
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4460795521736145,
      "eval_runtime": 0.6895,
      "eval_samples_per_second": 62.366,
      "eval_steps_per_second": 2.901,
      "step": 88
    },
    {
      "epoch": 45.0,
      "grad_norm": 3.4612958431243896,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.1459,
      "step": 90
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3602936565876007,
      "eval_runtime": 0.6778,
      "eval_samples_per_second": 63.442,
      "eval_steps_per_second": 2.951,
      "step": 90
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.29728585481643677,
      "eval_runtime": 0.6933,
      "eval_samples_per_second": 62.026,
      "eval_steps_per_second": 2.885,
      "step": 92
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.3384946584701538,
      "eval_runtime": 0.6497,
      "eval_samples_per_second": 66.182,
      "eval_steps_per_second": 3.078,
      "step": 94
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.3239303231239319,
      "eval_runtime": 0.6554,
      "eval_samples_per_second": 65.605,
      "eval_steps_per_second": 3.051,
      "step": 96
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4314607083797455,
      "eval_runtime": 0.6622,
      "eval_samples_per_second": 64.936,
      "eval_steps_per_second": 3.02,
      "step": 98
    },
    {
      "epoch": 50.0,
      "grad_norm": 5.4592132568359375,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1372,
      "step": 100
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.35193151235580444,
      "eval_runtime": 0.6733,
      "eval_samples_per_second": 63.868,
      "eval_steps_per_second": 2.971,
      "step": 100
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4147534966468811,
      "eval_runtime": 0.6694,
      "eval_samples_per_second": 64.238,
      "eval_steps_per_second": 2.988,
      "step": 102
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.46872177720069885,
      "eval_runtime": 0.6667,
      "eval_samples_per_second": 64.497,
      "eval_steps_per_second": 3.0,
      "step": 104
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.32870903611183167,
      "eval_runtime": 0.66,
      "eval_samples_per_second": 65.151,
      "eval_steps_per_second": 3.03,
      "step": 106
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.319438636302948,
      "eval_runtime": 0.6597,
      "eval_samples_per_second": 65.181,
      "eval_steps_per_second": 3.032,
      "step": 108
    },
    {
      "epoch": 55.0,
      "grad_norm": 4.405796527862549,
      "learning_rate": 2.5e-05,
      "loss": 0.1049,
      "step": 110
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.37025612592697144,
      "eval_runtime": 0.6736,
      "eval_samples_per_second": 63.836,
      "eval_steps_per_second": 2.969,
      "step": 110
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.35220110416412354,
      "eval_runtime": 0.673,
      "eval_samples_per_second": 63.894,
      "eval_steps_per_second": 2.972,
      "step": 112
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.2571551203727722,
      "eval_runtime": 0.6651,
      "eval_samples_per_second": 64.655,
      "eval_steps_per_second": 3.007,
      "step": 114
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.2522774338722229,
      "eval_runtime": 0.6708,
      "eval_samples_per_second": 64.099,
      "eval_steps_per_second": 2.981,
      "step": 116
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.313634991645813,
      "eval_runtime": 0.6692,
      "eval_samples_per_second": 64.252,
      "eval_steps_per_second": 2.988,
      "step": 118
    },
    {
      "epoch": 60.0,
      "grad_norm": 3.4014875888824463,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1143,
      "step": 120
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.36383214592933655,
      "eval_runtime": 0.669,
      "eval_samples_per_second": 64.278,
      "eval_steps_per_second": 2.99,
      "step": 120
    },
    {
      "epoch": 61.0,
      "eval_accuracy": 0.9534883720930233,
      "eval_loss": 0.29163503646850586,
      "eval_runtime": 0.6565,
      "eval_samples_per_second": 65.497,
      "eval_steps_per_second": 3.046,
      "step": 122
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.25211331248283386,
      "eval_runtime": 0.6574,
      "eval_samples_per_second": 65.406,
      "eval_steps_per_second": 3.042,
      "step": 124
    },
    {
      "epoch": 63.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.27347129583358765,
      "eval_runtime": 0.6629,
      "eval_samples_per_second": 64.871,
      "eval_steps_per_second": 3.017,
      "step": 126
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3111872375011444,
      "eval_runtime": 0.6634,
      "eval_samples_per_second": 64.822,
      "eval_steps_per_second": 3.015,
      "step": 128
    },
    {
      "epoch": 65.0,
      "grad_norm": 3.575437545776367,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.0885,
      "step": 130
    },
    {
      "epoch": 65.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3245609402656555,
      "eval_runtime": 0.6647,
      "eval_samples_per_second": 64.687,
      "eval_steps_per_second": 3.009,
      "step": 130
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.32640472054481506,
      "eval_runtime": 0.6683,
      "eval_samples_per_second": 64.343,
      "eval_steps_per_second": 2.993,
      "step": 132
    },
    {
      "epoch": 67.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.33514827489852905,
      "eval_runtime": 0.6574,
      "eval_samples_per_second": 65.406,
      "eval_steps_per_second": 3.042,
      "step": 134
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.34545841813087463,
      "eval_runtime": 0.6611,
      "eval_samples_per_second": 65.042,
      "eval_steps_per_second": 3.025,
      "step": 136
    },
    {
      "epoch": 69.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.35793864727020264,
      "eval_runtime": 0.6682,
      "eval_samples_per_second": 64.357,
      "eval_steps_per_second": 2.993,
      "step": 138
    },
    {
      "epoch": 70.0,
      "grad_norm": 4.83143424987793,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.1064,
      "step": 140
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3925873041152954,
      "eval_runtime": 0.6605,
      "eval_samples_per_second": 65.101,
      "eval_steps_per_second": 3.028,
      "step": 140
    },
    {
      "epoch": 71.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.43698573112487793,
      "eval_runtime": 0.6647,
      "eval_samples_per_second": 64.69,
      "eval_steps_per_second": 3.009,
      "step": 142
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.41486823558807373,
      "eval_runtime": 0.6615,
      "eval_samples_per_second": 65.007,
      "eval_steps_per_second": 3.024,
      "step": 144
    },
    {
      "epoch": 73.0,
      "eval_accuracy": 0.9534883720930233,
      "eval_loss": 0.3315199315547943,
      "eval_runtime": 0.6645,
      "eval_samples_per_second": 64.714,
      "eval_steps_per_second": 3.01,
      "step": 146
    },
    {
      "epoch": 74.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.270411878824234,
      "eval_runtime": 0.6615,
      "eval_samples_per_second": 65.002,
      "eval_steps_per_second": 3.023,
      "step": 148
    },
    {
      "epoch": 75.0,
      "grad_norm": 7.243945121765137,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.1047,
      "step": 150
    },
    {
      "epoch": 75.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.2600299119949341,
      "eval_runtime": 0.6636,
      "eval_samples_per_second": 64.798,
      "eval_steps_per_second": 3.014,
      "step": 150
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.9534883720930233,
      "eval_loss": 0.3215310871601105,
      "eval_runtime": 0.6604,
      "eval_samples_per_second": 65.115,
      "eval_steps_per_second": 3.029,
      "step": 152
    },
    {
      "epoch": 77.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.41095924377441406,
      "eval_runtime": 0.6667,
      "eval_samples_per_second": 64.5,
      "eval_steps_per_second": 3.0,
      "step": 154
    },
    {
      "epoch": 78.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4414052665233612,
      "eval_runtime": 0.6672,
      "eval_samples_per_second": 64.446,
      "eval_steps_per_second": 2.997,
      "step": 156
    },
    {
      "epoch": 79.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.35889923572540283,
      "eval_runtime": 0.6616,
      "eval_samples_per_second": 64.99,
      "eval_steps_per_second": 3.023,
      "step": 158
    },
    {
      "epoch": 80.0,
      "grad_norm": 4.898294925689697,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.0937,
      "step": 160
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.9534883720930233,
      "eval_loss": 0.3085266649723053,
      "eval_runtime": 0.6638,
      "eval_samples_per_second": 64.774,
      "eval_steps_per_second": 3.013,
      "step": 160
    },
    {
      "epoch": 81.0,
      "eval_accuracy": 0.9534883720930233,
      "eval_loss": 0.2889205813407898,
      "eval_runtime": 0.6641,
      "eval_samples_per_second": 64.752,
      "eval_steps_per_second": 3.012,
      "step": 162
    },
    {
      "epoch": 82.0,
      "eval_accuracy": 0.9534883720930233,
      "eval_loss": 0.2787427604198456,
      "eval_runtime": 0.6786,
      "eval_samples_per_second": 63.365,
      "eval_steps_per_second": 2.947,
      "step": 164
    },
    {
      "epoch": 83.0,
      "eval_accuracy": 0.9534883720930233,
      "eval_loss": 0.3251379728317261,
      "eval_runtime": 0.6658,
      "eval_samples_per_second": 64.584,
      "eval_steps_per_second": 3.004,
      "step": 166
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.44833165407180786,
      "eval_runtime": 0.6681,
      "eval_samples_per_second": 64.363,
      "eval_steps_per_second": 2.994,
      "step": 168
    },
    {
      "epoch": 85.0,
      "grad_norm": 3.011957883834839,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.0748,
      "step": 170
    },
    {
      "epoch": 85.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5489687919616699,
      "eval_runtime": 0.6674,
      "eval_samples_per_second": 64.431,
      "eval_steps_per_second": 2.997,
      "step": 170
    },
    {
      "epoch": 86.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5422279834747314,
      "eval_runtime": 0.6863,
      "eval_samples_per_second": 62.651,
      "eval_steps_per_second": 2.914,
      "step": 172
    },
    {
      "epoch": 87.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.5282132625579834,
      "eval_runtime": 0.6801,
      "eval_samples_per_second": 63.227,
      "eval_steps_per_second": 2.941,
      "step": 174
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5732578635215759,
      "eval_runtime": 0.6637,
      "eval_samples_per_second": 64.791,
      "eval_steps_per_second": 3.014,
      "step": 176
    },
    {
      "epoch": 89.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5977942943572998,
      "eval_runtime": 0.6668,
      "eval_samples_per_second": 64.486,
      "eval_steps_per_second": 2.999,
      "step": 178
    },
    {
      "epoch": 90.0,
      "grad_norm": 4.916757583618164,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0834,
      "step": 180
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5763372182846069,
      "eval_runtime": 0.6739,
      "eval_samples_per_second": 63.804,
      "eval_steps_per_second": 2.968,
      "step": 180
    },
    {
      "epoch": 91.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5270379185676575,
      "eval_runtime": 0.6653,
      "eval_samples_per_second": 64.633,
      "eval_steps_per_second": 3.006,
      "step": 182
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.49459221959114075,
      "eval_runtime": 0.6605,
      "eval_samples_per_second": 65.101,
      "eval_steps_per_second": 3.028,
      "step": 184
    },
    {
      "epoch": 93.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.48813045024871826,
      "eval_runtime": 0.6625,
      "eval_samples_per_second": 64.905,
      "eval_steps_per_second": 3.019,
      "step": 186
    },
    {
      "epoch": 94.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5114860534667969,
      "eval_runtime": 0.6698,
      "eval_samples_per_second": 64.203,
      "eval_steps_per_second": 2.986,
      "step": 188
    },
    {
      "epoch": 95.0,
      "grad_norm": 2.82369065284729,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.1016,
      "step": 190
    },
    {
      "epoch": 95.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5444539189338684,
      "eval_runtime": 0.6564,
      "eval_samples_per_second": 65.51,
      "eval_steps_per_second": 3.047,
      "step": 190
    },
    {
      "epoch": 96.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5536861419677734,
      "eval_runtime": 0.669,
      "eval_samples_per_second": 64.275,
      "eval_steps_per_second": 2.99,
      "step": 192
    },
    {
      "epoch": 97.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5450504422187805,
      "eval_runtime": 0.6786,
      "eval_samples_per_second": 63.366,
      "eval_steps_per_second": 2.947,
      "step": 194
    },
    {
      "epoch": 98.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5323299169540405,
      "eval_runtime": 0.6629,
      "eval_samples_per_second": 64.866,
      "eval_steps_per_second": 3.017,
      "step": 196
    },
    {
      "epoch": 99.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.5190114378929138,
      "eval_runtime": 0.6837,
      "eval_samples_per_second": 62.894,
      "eval_steps_per_second": 2.925,
      "step": 198
    },
    {
      "epoch": 100.0,
      "grad_norm": 4.195466041564941,
      "learning_rate": 0.0,
      "loss": 0.0657,
      "step": 200
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.5154587030410767,
      "eval_runtime": 0.661,
      "eval_samples_per_second": 65.052,
      "eval_steps_per_second": 3.026,
      "step": 200
    },
    {
      "epoch": 100.0,
      "step": 200,
      "total_flos": 1.8667587341684736e+18,
      "train_loss": 0.1868886011838913,
      "train_runtime": 1478.3752,
      "train_samples_per_second": 16.302,
      "train_steps_per_second": 0.135
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.9534883720930233,
      "eval_loss": 0.2663915455341339,
      "eval_runtime": 0.7149,
      "eval_samples_per_second": 60.151,
      "eval_steps_per_second": 2.798,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "total_flos": 1.8667587341684736e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}