{
  "best_metric": 0.9302325581395349,
  "best_model_checkpoint": "beit-base-patch16-224-75-fold1/checkpoint-68",
  "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": 0.3023255813953488,
      "eval_loss": 1.0986671447753906,
      "eval_runtime": 0.7121,
      "eval_samples_per_second": 60.387,
      "eval_steps_per_second": 2.809,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.6629528403282166,
      "eval_runtime": 0.6771,
      "eval_samples_per_second": 63.502,
      "eval_steps_per_second": 2.954,
      "step": 4
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.8342188596725464,
      "eval_runtime": 0.6723,
      "eval_samples_per_second": 63.959,
      "eval_steps_per_second": 2.975,
      "step": 6
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.6751894354820251,
      "eval_runtime": 0.6548,
      "eval_samples_per_second": 65.668,
      "eval_steps_per_second": 3.054,
      "step": 8
    },
    {
      "epoch": 5.0,
      "grad_norm": 11.474836349487305,
      "learning_rate": 2.5e-05,
      "loss": 0.7768,
      "step": 10
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.5407979488372803,
      "eval_runtime": 0.6694,
      "eval_samples_per_second": 64.236,
      "eval_steps_per_second": 2.988,
      "step": 10
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.725167453289032,
      "eval_runtime": 0.6793,
      "eval_samples_per_second": 63.302,
      "eval_steps_per_second": 2.944,
      "step": 12
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.7209302325581395,
      "eval_loss": 0.5609351396560669,
      "eval_runtime": 0.6818,
      "eval_samples_per_second": 63.069,
      "eval_steps_per_second": 2.933,
      "step": 14
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.7344595193862915,
      "eval_runtime": 0.6843,
      "eval_samples_per_second": 62.837,
      "eval_steps_per_second": 2.923,
      "step": 16
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.7674418604651163,
      "eval_loss": 0.461434930562973,
      "eval_runtime": 0.6823,
      "eval_samples_per_second": 63.024,
      "eval_steps_per_second": 2.931,
      "step": 18
    },
    {
      "epoch": 10.0,
      "grad_norm": 6.287548542022705,
      "learning_rate": 5e-05,
      "loss": 0.4249,
      "step": 20
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.4434371292591095,
      "eval_runtime": 0.6778,
      "eval_samples_per_second": 63.439,
      "eval_steps_per_second": 2.951,
      "step": 20
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.7551921606063843,
      "eval_runtime": 0.6784,
      "eval_samples_per_second": 63.381,
      "eval_steps_per_second": 2.948,
      "step": 22
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.7674418604651163,
      "eval_loss": 0.4141569435596466,
      "eval_runtime": 0.6771,
      "eval_samples_per_second": 63.504,
      "eval_steps_per_second": 2.954,
      "step": 24
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.7182537317276001,
      "eval_runtime": 0.6608,
      "eval_samples_per_second": 65.074,
      "eval_steps_per_second": 3.027,
      "step": 26
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.559093177318573,
      "eval_runtime": 0.6615,
      "eval_samples_per_second": 65.008,
      "eval_steps_per_second": 3.024,
      "step": 28
    },
    {
      "epoch": 15.0,
      "grad_norm": 13.554100036621094,
      "learning_rate": 4.722222222222222e-05,
      "loss": 0.3506,
      "step": 30
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.43631818890571594,
      "eval_runtime": 0.6576,
      "eval_samples_per_second": 65.39,
      "eval_steps_per_second": 3.041,
      "step": 30
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.5738235712051392,
      "eval_runtime": 0.663,
      "eval_samples_per_second": 64.853,
      "eval_steps_per_second": 3.016,
      "step": 32
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.4285888969898224,
      "eval_runtime": 0.6709,
      "eval_samples_per_second": 64.096,
      "eval_steps_per_second": 2.981,
      "step": 34
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.41996148228645325,
      "eval_runtime": 0.6718,
      "eval_samples_per_second": 64.007,
      "eval_steps_per_second": 2.977,
      "step": 36
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.6514244675636292,
      "eval_runtime": 0.6717,
      "eval_samples_per_second": 64.019,
      "eval_steps_per_second": 2.978,
      "step": 38
    },
    {
      "epoch": 20.0,
      "grad_norm": 3.562516212463379,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.3434,
      "step": 40
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.41895100474357605,
      "eval_runtime": 0.6599,
      "eval_samples_per_second": 65.159,
      "eval_steps_per_second": 3.031,
      "step": 40
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.6220327615737915,
      "eval_runtime": 0.6586,
      "eval_samples_per_second": 65.286,
      "eval_steps_per_second": 3.037,
      "step": 42
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.6333664059638977,
      "eval_runtime": 0.6711,
      "eval_samples_per_second": 64.071,
      "eval_steps_per_second": 2.98,
      "step": 44
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.44870051741600037,
      "eval_runtime": 0.6629,
      "eval_samples_per_second": 64.869,
      "eval_steps_per_second": 3.017,
      "step": 46
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.49601471424102783,
      "eval_runtime": 0.6611,
      "eval_samples_per_second": 65.043,
      "eval_steps_per_second": 3.025,
      "step": 48
    },
    {
      "epoch": 25.0,
      "grad_norm": 10.35086441040039,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.2498,
      "step": 50
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.4178723990917206,
      "eval_runtime": 0.6622,
      "eval_samples_per_second": 64.938,
      "eval_steps_per_second": 3.02,
      "step": 50
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.32208359241485596,
      "eval_runtime": 0.6585,
      "eval_samples_per_second": 65.297,
      "eval_steps_per_second": 3.037,
      "step": 52
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.4776231646537781,
      "eval_runtime": 0.6578,
      "eval_samples_per_second": 65.371,
      "eval_steps_per_second": 3.041,
      "step": 54
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.5755565762519836,
      "eval_runtime": 0.6653,
      "eval_samples_per_second": 64.629,
      "eval_steps_per_second": 3.006,
      "step": 56
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.5443851351737976,
      "eval_runtime": 0.6687,
      "eval_samples_per_second": 64.308,
      "eval_steps_per_second": 2.991,
      "step": 58
    },
    {
      "epoch": 30.0,
      "grad_norm": 6.6725850105285645,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.2461,
      "step": 60
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.39731207489967346,
      "eval_runtime": 0.6614,
      "eval_samples_per_second": 65.018,
      "eval_steps_per_second": 3.024,
      "step": 60
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.36715105175971985,
      "eval_runtime": 0.6663,
      "eval_samples_per_second": 64.533,
      "eval_steps_per_second": 3.002,
      "step": 62
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.40706878900527954,
      "eval_runtime": 0.6733,
      "eval_samples_per_second": 63.866,
      "eval_steps_per_second": 2.971,
      "step": 64
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.7674418604651163,
      "eval_loss": 0.46778932213783264,
      "eval_runtime": 0.6604,
      "eval_samples_per_second": 65.109,
      "eval_steps_per_second": 3.028,
      "step": 66
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.26408103108406067,
      "eval_runtime": 0.6719,
      "eval_samples_per_second": 63.996,
      "eval_steps_per_second": 2.977,
      "step": 68
    },
    {
      "epoch": 35.0,
      "grad_norm": 8.221508979797363,
      "learning_rate": 3.611111111111111e-05,
      "loss": 0.2279,
      "step": 70
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.5550601482391357,
      "eval_runtime": 0.6604,
      "eval_samples_per_second": 65.113,
      "eval_steps_per_second": 3.029,
      "step": 70
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.2726641893386841,
      "eval_runtime": 0.6619,
      "eval_samples_per_second": 64.963,
      "eval_steps_per_second": 3.022,
      "step": 72
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.33118677139282227,
      "eval_runtime": 0.6645,
      "eval_samples_per_second": 64.71,
      "eval_steps_per_second": 3.01,
      "step": 74
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.748543381690979,
      "eval_runtime": 0.6686,
      "eval_samples_per_second": 64.315,
      "eval_steps_per_second": 2.991,
      "step": 76
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.6406599283218384,
      "eval_runtime": 0.6631,
      "eval_samples_per_second": 64.847,
      "eval_steps_per_second": 3.016,
      "step": 78
    },
    {
      "epoch": 40.0,
      "grad_norm": 5.549210071563721,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.183,
      "step": 80
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.5420320630073547,
      "eval_runtime": 0.6604,
      "eval_samples_per_second": 65.113,
      "eval_steps_per_second": 3.029,
      "step": 80
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.736416220664978,
      "eval_runtime": 0.6634,
      "eval_samples_per_second": 64.813,
      "eval_steps_per_second": 3.015,
      "step": 82
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.4140843152999878,
      "eval_runtime": 0.6681,
      "eval_samples_per_second": 64.364,
      "eval_steps_per_second": 2.994,
      "step": 84
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.5461295247077942,
      "eval_runtime": 0.6611,
      "eval_samples_per_second": 65.046,
      "eval_steps_per_second": 3.025,
      "step": 86
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.3437821567058563,
      "eval_runtime": 0.6752,
      "eval_samples_per_second": 63.684,
      "eval_steps_per_second": 2.962,
      "step": 88
    },
    {
      "epoch": 45.0,
      "grad_norm": 6.833462715148926,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.1658,
      "step": 90
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3321613669395447,
      "eval_runtime": 0.6603,
      "eval_samples_per_second": 65.118,
      "eval_steps_per_second": 3.029,
      "step": 90
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3463155925273895,
      "eval_runtime": 0.6639,
      "eval_samples_per_second": 64.772,
      "eval_steps_per_second": 3.013,
      "step": 92
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.606645941734314,
      "eval_runtime": 0.6821,
      "eval_samples_per_second": 63.037,
      "eval_steps_per_second": 2.932,
      "step": 94
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.6258805990219116,
      "eval_runtime": 0.6758,
      "eval_samples_per_second": 63.633,
      "eval_steps_per_second": 2.96,
      "step": 96
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.49087798595428467,
      "eval_runtime": 0.6712,
      "eval_samples_per_second": 64.069,
      "eval_steps_per_second": 2.98,
      "step": 98
    },
    {
      "epoch": 50.0,
      "grad_norm": 5.422869682312012,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1555,
      "step": 100
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.7906976744186046,
      "eval_loss": 0.6021594405174255,
      "eval_runtime": 0.6752,
      "eval_samples_per_second": 63.682,
      "eval_steps_per_second": 2.962,
      "step": 100
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.5234286189079285,
      "eval_runtime": 0.6625,
      "eval_samples_per_second": 64.902,
      "eval_steps_per_second": 3.019,
      "step": 102
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.416355699300766,
      "eval_runtime": 0.6553,
      "eval_samples_per_second": 65.615,
      "eval_steps_per_second": 3.052,
      "step": 104
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.3893319070339203,
      "eval_runtime": 0.6683,
      "eval_samples_per_second": 64.341,
      "eval_steps_per_second": 2.993,
      "step": 106
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.3773750364780426,
      "eval_runtime": 0.6579,
      "eval_samples_per_second": 65.363,
      "eval_steps_per_second": 3.04,
      "step": 108
    },
    {
      "epoch": 55.0,
      "grad_norm": 3.3167049884796143,
      "learning_rate": 2.5e-05,
      "loss": 0.1487,
      "step": 110
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.7532004714012146,
      "eval_runtime": 0.6623,
      "eval_samples_per_second": 64.926,
      "eval_steps_per_second": 3.02,
      "step": 110
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.7141379714012146,
      "eval_runtime": 0.6724,
      "eval_samples_per_second": 63.949,
      "eval_steps_per_second": 2.974,
      "step": 112
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.41971808671951294,
      "eval_runtime": 0.6716,
      "eval_samples_per_second": 64.029,
      "eval_steps_per_second": 2.978,
      "step": 114
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.7441860465116279,
      "eval_loss": 0.6815913915634155,
      "eval_runtime": 0.669,
      "eval_samples_per_second": 64.273,
      "eval_steps_per_second": 2.989,
      "step": 116
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5383610725402832,
      "eval_runtime": 0.6714,
      "eval_samples_per_second": 64.049,
      "eval_steps_per_second": 2.979,
      "step": 118
    },
    {
      "epoch": 60.0,
      "grad_norm": 4.153010845184326,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.1349,
      "step": 120
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.4970913231372833,
      "eval_runtime": 0.6622,
      "eval_samples_per_second": 64.939,
      "eval_steps_per_second": 3.02,
      "step": 120
    },
    {
      "epoch": 61.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.46013569831848145,
      "eval_runtime": 0.6877,
      "eval_samples_per_second": 62.527,
      "eval_steps_per_second": 2.908,
      "step": 122
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.4739607870578766,
      "eval_runtime": 0.7045,
      "eval_samples_per_second": 61.038,
      "eval_steps_per_second": 2.839,
      "step": 124
    },
    {
      "epoch": 63.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5385664701461792,
      "eval_runtime": 0.6901,
      "eval_samples_per_second": 62.306,
      "eval_steps_per_second": 2.898,
      "step": 126
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.337642639875412,
      "eval_runtime": 0.6889,
      "eval_samples_per_second": 62.417,
      "eval_steps_per_second": 2.903,
      "step": 128
    },
    {
      "epoch": 65.0,
      "grad_norm": 2.339792013168335,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.128,
      "step": 130
    },
    {
      "epoch": 65.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3904857933521271,
      "eval_runtime": 0.7005,
      "eval_samples_per_second": 61.389,
      "eval_steps_per_second": 2.855,
      "step": 130
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.38408252596855164,
      "eval_runtime": 0.6804,
      "eval_samples_per_second": 63.2,
      "eval_steps_per_second": 2.94,
      "step": 132
    },
    {
      "epoch": 67.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.35673901438713074,
      "eval_runtime": 0.6679,
      "eval_samples_per_second": 64.376,
      "eval_steps_per_second": 2.994,
      "step": 134
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.3984518349170685,
      "eval_runtime": 0.6713,
      "eval_samples_per_second": 64.054,
      "eval_steps_per_second": 2.979,
      "step": 136
    },
    {
      "epoch": 69.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.41648656129837036,
      "eval_runtime": 0.6648,
      "eval_samples_per_second": 64.679,
      "eval_steps_per_second": 3.008,
      "step": 138
    },
    {
      "epoch": 70.0,
      "grad_norm": 1.7775969505310059,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.0875,
      "step": 140
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.4346034526824951,
      "eval_runtime": 0.6646,
      "eval_samples_per_second": 64.699,
      "eval_steps_per_second": 3.009,
      "step": 140
    },
    {
      "epoch": 71.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.44971972703933716,
      "eval_runtime": 0.664,
      "eval_samples_per_second": 64.756,
      "eval_steps_per_second": 3.012,
      "step": 142
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4352738559246063,
      "eval_runtime": 0.6621,
      "eval_samples_per_second": 64.948,
      "eval_steps_per_second": 3.021,
      "step": 144
    },
    {
      "epoch": 73.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.42759430408477783,
      "eval_runtime": 0.6825,
      "eval_samples_per_second": 63.008,
      "eval_steps_per_second": 2.931,
      "step": 146
    },
    {
      "epoch": 74.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4010199308395386,
      "eval_runtime": 0.6666,
      "eval_samples_per_second": 64.502,
      "eval_steps_per_second": 3.0,
      "step": 148
    },
    {
      "epoch": 75.0,
      "grad_norm": 3.5295796394348145,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.0932,
      "step": 150
    },
    {
      "epoch": 75.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.39583244919776917,
      "eval_runtime": 0.6712,
      "eval_samples_per_second": 64.065,
      "eval_steps_per_second": 2.98,
      "step": 150
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3603510558605194,
      "eval_runtime": 0.6628,
      "eval_samples_per_second": 64.873,
      "eval_steps_per_second": 3.017,
      "step": 152
    },
    {
      "epoch": 77.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.3426762819290161,
      "eval_runtime": 0.6836,
      "eval_samples_per_second": 62.899,
      "eval_steps_per_second": 2.926,
      "step": 154
    },
    {
      "epoch": 78.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.3416913151741028,
      "eval_runtime": 0.6593,
      "eval_samples_per_second": 65.218,
      "eval_steps_per_second": 3.033,
      "step": 156
    },
    {
      "epoch": 79.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3438415229320526,
      "eval_runtime": 0.6796,
      "eval_samples_per_second": 63.275,
      "eval_steps_per_second": 2.943,
      "step": 158
    },
    {
      "epoch": 80.0,
      "grad_norm": 3.4679925441741943,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.0943,
      "step": 160
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.37560805678367615,
      "eval_runtime": 0.6668,
      "eval_samples_per_second": 64.49,
      "eval_steps_per_second": 3.0,
      "step": 160
    },
    {
      "epoch": 81.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.40767714381217957,
      "eval_runtime": 0.67,
      "eval_samples_per_second": 64.184,
      "eval_steps_per_second": 2.985,
      "step": 162
    },
    {
      "epoch": 82.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.4129002094268799,
      "eval_runtime": 0.6659,
      "eval_samples_per_second": 64.577,
      "eval_steps_per_second": 3.004,
      "step": 164
    },
    {
      "epoch": 83.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.43040457367897034,
      "eval_runtime": 0.6583,
      "eval_samples_per_second": 65.322,
      "eval_steps_per_second": 3.038,
      "step": 166
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.41560518741607666,
      "eval_runtime": 0.6588,
      "eval_samples_per_second": 65.272,
      "eval_steps_per_second": 3.036,
      "step": 168
    },
    {
      "epoch": 85.0,
      "grad_norm": 3.2857208251953125,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.0753,
      "step": 170
    },
    {
      "epoch": 85.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.40876707434654236,
      "eval_runtime": 0.664,
      "eval_samples_per_second": 64.76,
      "eval_steps_per_second": 3.012,
      "step": 170
    },
    {
      "epoch": 86.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4090438187122345,
      "eval_runtime": 0.6671,
      "eval_samples_per_second": 64.46,
      "eval_steps_per_second": 2.998,
      "step": 172
    },
    {
      "epoch": 87.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.40758630633354187,
      "eval_runtime": 0.6654,
      "eval_samples_per_second": 64.62,
      "eval_steps_per_second": 3.006,
      "step": 174
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.42727890610694885,
      "eval_runtime": 0.6624,
      "eval_samples_per_second": 64.913,
      "eval_steps_per_second": 3.019,
      "step": 176
    },
    {
      "epoch": 89.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.43666666746139526,
      "eval_runtime": 0.6738,
      "eval_samples_per_second": 63.814,
      "eval_steps_per_second": 2.968,
      "step": 178
    },
    {
      "epoch": 90.0,
      "grad_norm": 3.541504144668579,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0846,
      "step": 180
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.4489597976207733,
      "eval_runtime": 0.658,
      "eval_samples_per_second": 65.353,
      "eval_steps_per_second": 3.04,
      "step": 180
    },
    {
      "epoch": 91.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4448123276233673,
      "eval_runtime": 0.6618,
      "eval_samples_per_second": 64.971,
      "eval_steps_per_second": 3.022,
      "step": 182
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4405640959739685,
      "eval_runtime": 0.6663,
      "eval_samples_per_second": 64.534,
      "eval_steps_per_second": 3.002,
      "step": 184
    },
    {
      "epoch": 93.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4392721951007843,
      "eval_runtime": 0.6649,
      "eval_samples_per_second": 64.673,
      "eval_steps_per_second": 3.008,
      "step": 186
    },
    {
      "epoch": 94.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4370172321796417,
      "eval_runtime": 0.662,
      "eval_samples_per_second": 64.953,
      "eval_steps_per_second": 3.021,
      "step": 188
    },
    {
      "epoch": 95.0,
      "grad_norm": 3.8935422897338867,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.0865,
      "step": 190
    },
    {
      "epoch": 95.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4330422580242157,
      "eval_runtime": 0.6653,
      "eval_samples_per_second": 64.633,
      "eval_steps_per_second": 3.006,
      "step": 190
    },
    {
      "epoch": 96.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.42933788895606995,
      "eval_runtime": 0.6605,
      "eval_samples_per_second": 65.104,
      "eval_steps_per_second": 3.028,
      "step": 192
    },
    {
      "epoch": 97.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4240441918373108,
      "eval_runtime": 0.6629,
      "eval_samples_per_second": 64.862,
      "eval_steps_per_second": 3.017,
      "step": 194
    },
    {
      "epoch": 98.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4176953434944153,
      "eval_runtime": 0.6716,
      "eval_samples_per_second": 64.025,
      "eval_steps_per_second": 2.978,
      "step": 196
    },
    {
      "epoch": 99.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4143863618373871,
      "eval_runtime": 0.6706,
      "eval_samples_per_second": 64.123,
      "eval_steps_per_second": 2.982,
      "step": 198
    },
    {
      "epoch": 100.0,
      "grad_norm": 2.7952256202697754,
      "learning_rate": 0.0,
      "loss": 0.1019,
      "step": 200
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4134748578071594,
      "eval_runtime": 0.6659,
      "eval_samples_per_second": 64.576,
      "eval_steps_per_second": 3.004,
      "step": 200
    },
    {
      "epoch": 100.0,
      "step": 200,
      "total_flos": 1.8667587341684736e+18,
      "train_loss": 0.20794262617826462,
      "train_runtime": 1197.3367,
      "train_samples_per_second": 20.128,
      "train_steps_per_second": 0.167
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.26408103108406067,
      "eval_runtime": 0.7086,
      "eval_samples_per_second": 60.684,
      "eval_steps_per_second": 2.823,
      "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
}