{
  "best_metric": 0.9302325581395349,
  "best_model_checkpoint": "deit-base-distilled-patch16-224-75-fold4/checkpoint-60",
  "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": 0.6744186046511628,
      "eval_loss": 0.5612571835517883,
      "eval_runtime": 0.6247,
      "eval_samples_per_second": 68.834,
      "eval_steps_per_second": 3.202,
      "step": 2
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.6232779622077942,
      "eval_runtime": 0.6154,
      "eval_samples_per_second": 69.878,
      "eval_steps_per_second": 3.25,
      "step": 4
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.7506104111671448,
      "eval_runtime": 0.5904,
      "eval_samples_per_second": 72.836,
      "eval_steps_per_second": 3.388,
      "step": 6
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.6976744186046512,
      "eval_loss": 0.646221399307251,
      "eval_runtime": 0.5997,
      "eval_samples_per_second": 71.707,
      "eval_steps_per_second": 3.335,
      "step": 8
    },
    {
      "epoch": 5.0,
      "grad_norm": 2.89682674407959,
      "learning_rate": 2.5e-05,
      "loss": 0.5231,
      "step": 10
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.388916939496994,
      "eval_runtime": 0.5966,
      "eval_samples_per_second": 72.074,
      "eval_steps_per_second": 3.352,
      "step": 10
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.34845367074012756,
      "eval_runtime": 0.6551,
      "eval_samples_per_second": 65.638,
      "eval_steps_per_second": 3.053,
      "step": 12
    },
    {
      "epoch": 7.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.4326934516429901,
      "eval_runtime": 0.6081,
      "eval_samples_per_second": 70.712,
      "eval_steps_per_second": 3.289,
      "step": 14
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.44700080156326294,
      "eval_runtime": 0.6098,
      "eval_samples_per_second": 70.519,
      "eval_steps_per_second": 3.28,
      "step": 16
    },
    {
      "epoch": 9.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.30670422315597534,
      "eval_runtime": 0.6047,
      "eval_samples_per_second": 71.107,
      "eval_steps_per_second": 3.307,
      "step": 18
    },
    {
      "epoch": 10.0,
      "grad_norm": 5.931050777435303,
      "learning_rate": 5e-05,
      "loss": 0.3699,
      "step": 20
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.30745768547058105,
      "eval_runtime": 0.6121,
      "eval_samples_per_second": 70.254,
      "eval_steps_per_second": 3.268,
      "step": 20
    },
    {
      "epoch": 11.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.5433412790298462,
      "eval_runtime": 0.569,
      "eval_samples_per_second": 75.57,
      "eval_steps_per_second": 3.515,
      "step": 22
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5848782658576965,
      "eval_runtime": 0.6053,
      "eval_samples_per_second": 71.041,
      "eval_steps_per_second": 3.304,
      "step": 24
    },
    {
      "epoch": 13.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.45092493295669556,
      "eval_runtime": 0.6096,
      "eval_samples_per_second": 70.539,
      "eval_steps_per_second": 3.281,
      "step": 26
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.5410060286521912,
      "eval_runtime": 0.6078,
      "eval_samples_per_second": 70.743,
      "eval_steps_per_second": 3.29,
      "step": 28
    },
    {
      "epoch": 15.0,
      "grad_norm": 1.49797785282135,
      "learning_rate": 4.722222222222222e-05,
      "loss": 0.2951,
      "step": 30
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.4709995985031128,
      "eval_runtime": 0.5995,
      "eval_samples_per_second": 71.724,
      "eval_steps_per_second": 3.336,
      "step": 30
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.37380093336105347,
      "eval_runtime": 0.6173,
      "eval_samples_per_second": 69.661,
      "eval_steps_per_second": 3.24,
      "step": 32
    },
    {
      "epoch": 17.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.3704395890235901,
      "eval_runtime": 0.6096,
      "eval_samples_per_second": 70.533,
      "eval_steps_per_second": 3.281,
      "step": 34
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.2682153284549713,
      "eval_runtime": 0.6298,
      "eval_samples_per_second": 68.28,
      "eval_steps_per_second": 3.176,
      "step": 36
    },
    {
      "epoch": 19.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.4187103509902954,
      "eval_runtime": 0.6071,
      "eval_samples_per_second": 70.824,
      "eval_steps_per_second": 3.294,
      "step": 38
    },
    {
      "epoch": 20.0,
      "grad_norm": 4.724278926849365,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.2279,
      "step": 40
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.39605996012687683,
      "eval_runtime": 0.6633,
      "eval_samples_per_second": 64.826,
      "eval_steps_per_second": 3.015,
      "step": 40
    },
    {
      "epoch": 21.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.29759395122528076,
      "eval_runtime": 0.6163,
      "eval_samples_per_second": 69.776,
      "eval_steps_per_second": 3.245,
      "step": 42
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.5654792189598083,
      "eval_runtime": 0.6301,
      "eval_samples_per_second": 68.243,
      "eval_steps_per_second": 3.174,
      "step": 44
    },
    {
      "epoch": 23.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.49307170510292053,
      "eval_runtime": 0.6365,
      "eval_samples_per_second": 67.56,
      "eval_steps_per_second": 3.142,
      "step": 46
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.2672000229358673,
      "eval_runtime": 0.636,
      "eval_samples_per_second": 67.61,
      "eval_steps_per_second": 3.145,
      "step": 48
    },
    {
      "epoch": 25.0,
      "grad_norm": 3.272207021713257,
      "learning_rate": 4.166666666666667e-05,
      "loss": 0.1823,
      "step": 50
    },
    {
      "epoch": 25.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.27949997782707214,
      "eval_runtime": 0.6181,
      "eval_samples_per_second": 69.573,
      "eval_steps_per_second": 3.236,
      "step": 50
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.38846340775489807,
      "eval_runtime": 0.6434,
      "eval_samples_per_second": 66.837,
      "eval_steps_per_second": 3.109,
      "step": 52
    },
    {
      "epoch": 27.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.27794820070266724,
      "eval_runtime": 0.6135,
      "eval_samples_per_second": 70.09,
      "eval_steps_per_second": 3.26,
      "step": 54
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.29900121688842773,
      "eval_runtime": 0.6165,
      "eval_samples_per_second": 69.752,
      "eval_steps_per_second": 3.244,
      "step": 56
    },
    {
      "epoch": 29.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.3090374171733856,
      "eval_runtime": 0.6066,
      "eval_samples_per_second": 70.883,
      "eval_steps_per_second": 3.297,
      "step": 58
    },
    {
      "epoch": 30.0,
      "grad_norm": 3.426194190979004,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.1314,
      "step": 60
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.25641849637031555,
      "eval_runtime": 0.6196,
      "eval_samples_per_second": 69.398,
      "eval_steps_per_second": 3.228,
      "step": 60
    },
    {
      "epoch": 31.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.28250229358673096,
      "eval_runtime": 0.6098,
      "eval_samples_per_second": 70.512,
      "eval_steps_per_second": 3.28,
      "step": 62
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.36547577381134033,
      "eval_runtime": 0.6196,
      "eval_samples_per_second": 69.405,
      "eval_steps_per_second": 3.228,
      "step": 64
    },
    {
      "epoch": 33.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.24863988161087036,
      "eval_runtime": 0.6058,
      "eval_samples_per_second": 70.978,
      "eval_steps_per_second": 3.301,
      "step": 66
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.2734163999557495,
      "eval_runtime": 0.6179,
      "eval_samples_per_second": 69.592,
      "eval_steps_per_second": 3.237,
      "step": 68
    },
    {
      "epoch": 35.0,
      "grad_norm": 2.8588931560516357,
      "learning_rate": 3.611111111111111e-05,
      "loss": 0.1387,
      "step": 70
    },
    {
      "epoch": 35.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.5241507291793823,
      "eval_runtime": 0.6101,
      "eval_samples_per_second": 70.484,
      "eval_steps_per_second": 3.278,
      "step": 70
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.5958200693130493,
      "eval_runtime": 0.6269,
      "eval_samples_per_second": 68.588,
      "eval_steps_per_second": 3.19,
      "step": 72
    },
    {
      "epoch": 37.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.34717297554016113,
      "eval_runtime": 0.6204,
      "eval_samples_per_second": 69.309,
      "eval_steps_per_second": 3.224,
      "step": 74
    },
    {
      "epoch": 38.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3057456314563751,
      "eval_runtime": 0.6213,
      "eval_samples_per_second": 69.211,
      "eval_steps_per_second": 3.219,
      "step": 76
    },
    {
      "epoch": 39.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.40181106328964233,
      "eval_runtime": 0.6158,
      "eval_samples_per_second": 69.825,
      "eval_steps_per_second": 3.248,
      "step": 78
    },
    {
      "epoch": 40.0,
      "grad_norm": 2.626035213470459,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.1266,
      "step": 80
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.8372093023255814,
      "eval_loss": 0.39403417706489563,
      "eval_runtime": 0.6245,
      "eval_samples_per_second": 68.855,
      "eval_steps_per_second": 3.203,
      "step": 80
    },
    {
      "epoch": 41.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.35684317350387573,
      "eval_runtime": 0.611,
      "eval_samples_per_second": 70.379,
      "eval_steps_per_second": 3.273,
      "step": 82
    },
    {
      "epoch": 42.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.35330983996391296,
      "eval_runtime": 0.6245,
      "eval_samples_per_second": 68.858,
      "eval_steps_per_second": 3.203,
      "step": 84
    },
    {
      "epoch": 43.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.34508436918258667,
      "eval_runtime": 0.6115,
      "eval_samples_per_second": 70.324,
      "eval_steps_per_second": 3.271,
      "step": 86
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.34777510166168213,
      "eval_runtime": 0.6166,
      "eval_samples_per_second": 69.739,
      "eval_steps_per_second": 3.244,
      "step": 88
    },
    {
      "epoch": 45.0,
      "grad_norm": 2.9467356204986572,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.1004,
      "step": 90
    },
    {
      "epoch": 45.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.31948891282081604,
      "eval_runtime": 0.6101,
      "eval_samples_per_second": 70.485,
      "eval_steps_per_second": 3.278,
      "step": 90
    },
    {
      "epoch": 46.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3925861418247223,
      "eval_runtime": 0.6225,
      "eval_samples_per_second": 69.08,
      "eval_steps_per_second": 3.213,
      "step": 92
    },
    {
      "epoch": 47.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4168529212474823,
      "eval_runtime": 0.6336,
      "eval_samples_per_second": 67.87,
      "eval_steps_per_second": 3.157,
      "step": 94
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.42742013931274414,
      "eval_runtime": 0.6433,
      "eval_samples_per_second": 66.848,
      "eval_steps_per_second": 3.109,
      "step": 96
    },
    {
      "epoch": 49.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.406142920255661,
      "eval_runtime": 0.6121,
      "eval_samples_per_second": 70.25,
      "eval_steps_per_second": 3.267,
      "step": 98
    },
    {
      "epoch": 50.0,
      "grad_norm": 2.1352977752685547,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.1033,
      "step": 100
    },
    {
      "epoch": 50.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.4276573657989502,
      "eval_runtime": 0.6177,
      "eval_samples_per_second": 69.616,
      "eval_steps_per_second": 3.238,
      "step": 100
    },
    {
      "epoch": 51.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.39769700169563293,
      "eval_runtime": 0.619,
      "eval_samples_per_second": 69.469,
      "eval_steps_per_second": 3.231,
      "step": 102
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.44276344776153564,
      "eval_runtime": 0.6222,
      "eval_samples_per_second": 69.111,
      "eval_steps_per_second": 3.214,
      "step": 104
    },
    {
      "epoch": 53.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.67534339427948,
      "eval_runtime": 0.6176,
      "eval_samples_per_second": 69.619,
      "eval_steps_per_second": 3.238,
      "step": 106
    },
    {
      "epoch": 54.0,
      "eval_accuracy": 0.813953488372093,
      "eval_loss": 0.6911506056785583,
      "eval_runtime": 0.6198,
      "eval_samples_per_second": 69.374,
      "eval_steps_per_second": 3.227,
      "step": 108
    },
    {
      "epoch": 55.0,
      "grad_norm": 1.6982649564743042,
      "learning_rate": 2.5e-05,
      "loss": 0.0827,
      "step": 110
    },
    {
      "epoch": 55.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.42013683915138245,
      "eval_runtime": 0.6111,
      "eval_samples_per_second": 70.369,
      "eval_steps_per_second": 3.273,
      "step": 110
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3086050748825073,
      "eval_runtime": 0.6171,
      "eval_samples_per_second": 69.676,
      "eval_steps_per_second": 3.241,
      "step": 112
    },
    {
      "epoch": 57.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.31496259570121765,
      "eval_runtime": 0.6141,
      "eval_samples_per_second": 70.016,
      "eval_steps_per_second": 3.257,
      "step": 114
    },
    {
      "epoch": 58.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.4757108986377716,
      "eval_runtime": 0.6178,
      "eval_samples_per_second": 69.6,
      "eval_steps_per_second": 3.237,
      "step": 116
    },
    {
      "epoch": 59.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.6409224271774292,
      "eval_runtime": 0.6132,
      "eval_samples_per_second": 70.118,
      "eval_steps_per_second": 3.261,
      "step": 118
    },
    {
      "epoch": 60.0,
      "grad_norm": 1.7638839483261108,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.0889,
      "step": 120
    },
    {
      "epoch": 60.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.5430170297622681,
      "eval_runtime": 0.6138,
      "eval_samples_per_second": 70.052,
      "eval_steps_per_second": 3.258,
      "step": 120
    },
    {
      "epoch": 61.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.404401570558548,
      "eval_runtime": 0.6134,
      "eval_samples_per_second": 70.099,
      "eval_steps_per_second": 3.26,
      "step": 122
    },
    {
      "epoch": 62.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.34726735949516296,
      "eval_runtime": 0.6164,
      "eval_samples_per_second": 69.76,
      "eval_steps_per_second": 3.245,
      "step": 124
    },
    {
      "epoch": 63.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3485052287578583,
      "eval_runtime": 0.6142,
      "eval_samples_per_second": 70.005,
      "eval_steps_per_second": 3.256,
      "step": 126
    },
    {
      "epoch": 64.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3711298108100891,
      "eval_runtime": 0.6207,
      "eval_samples_per_second": 69.276,
      "eval_steps_per_second": 3.222,
      "step": 128
    },
    {
      "epoch": 65.0,
      "grad_norm": 2.3730201721191406,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.088,
      "step": 130
    },
    {
      "epoch": 65.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4405280649662018,
      "eval_runtime": 0.6058,
      "eval_samples_per_second": 70.975,
      "eval_steps_per_second": 3.301,
      "step": 130
    },
    {
      "epoch": 66.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.6526488065719604,
      "eval_runtime": 0.6239,
      "eval_samples_per_second": 68.921,
      "eval_steps_per_second": 3.206,
      "step": 132
    },
    {
      "epoch": 67.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.701919674873352,
      "eval_runtime": 0.6164,
      "eval_samples_per_second": 69.76,
      "eval_steps_per_second": 3.245,
      "step": 134
    },
    {
      "epoch": 68.0,
      "eval_accuracy": 0.8604651162790697,
      "eval_loss": 0.540764570236206,
      "eval_runtime": 0.6107,
      "eval_samples_per_second": 70.416,
      "eval_steps_per_second": 3.275,
      "step": 136
    },
    {
      "epoch": 69.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.4056636393070221,
      "eval_runtime": 0.6116,
      "eval_samples_per_second": 70.303,
      "eval_steps_per_second": 3.27,
      "step": 138
    },
    {
      "epoch": 70.0,
      "grad_norm": 3.548440933227539,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.0734,
      "step": 140
    },
    {
      "epoch": 70.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.37973058223724365,
      "eval_runtime": 0.6421,
      "eval_samples_per_second": 66.965,
      "eval_steps_per_second": 3.115,
      "step": 140
    },
    {
      "epoch": 71.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3811149299144745,
      "eval_runtime": 0.6319,
      "eval_samples_per_second": 68.049,
      "eval_steps_per_second": 3.165,
      "step": 142
    },
    {
      "epoch": 72.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.4039730727672577,
      "eval_runtime": 0.614,
      "eval_samples_per_second": 70.03,
      "eval_steps_per_second": 3.257,
      "step": 144
    },
    {
      "epoch": 73.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4566517770290375,
      "eval_runtime": 0.6134,
      "eval_samples_per_second": 70.099,
      "eval_steps_per_second": 3.26,
      "step": 146
    },
    {
      "epoch": 74.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.5160852074623108,
      "eval_runtime": 0.623,
      "eval_samples_per_second": 69.016,
      "eval_steps_per_second": 3.21,
      "step": 148
    },
    {
      "epoch": 75.0,
      "grad_norm": 1.0712120532989502,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.0721,
      "step": 150
    },
    {
      "epoch": 75.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.5240259766578674,
      "eval_runtime": 0.6079,
      "eval_samples_per_second": 70.737,
      "eval_steps_per_second": 3.29,
      "step": 150
    },
    {
      "epoch": 76.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.5048332214355469,
      "eval_runtime": 0.6233,
      "eval_samples_per_second": 68.988,
      "eval_steps_per_second": 3.209,
      "step": 152
    },
    {
      "epoch": 77.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.4634605348110199,
      "eval_runtime": 0.611,
      "eval_samples_per_second": 70.376,
      "eval_steps_per_second": 3.273,
      "step": 154
    },
    {
      "epoch": 78.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.45102155208587646,
      "eval_runtime": 0.6381,
      "eval_samples_per_second": 67.39,
      "eval_steps_per_second": 3.134,
      "step": 156
    },
    {
      "epoch": 79.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.49311599135398865,
      "eval_runtime": 0.6161,
      "eval_samples_per_second": 69.791,
      "eval_steps_per_second": 3.246,
      "step": 158
    },
    {
      "epoch": 80.0,
      "grad_norm": 1.1070908308029175,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.0592,
      "step": 160
    },
    {
      "epoch": 80.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.5368152856826782,
      "eval_runtime": 0.6099,
      "eval_samples_per_second": 70.503,
      "eval_steps_per_second": 3.279,
      "step": 160
    },
    {
      "epoch": 81.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.5296711921691895,
      "eval_runtime": 0.6174,
      "eval_samples_per_second": 69.644,
      "eval_steps_per_second": 3.239,
      "step": 162
    },
    {
      "epoch": 82.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.4721624255180359,
      "eval_runtime": 0.62,
      "eval_samples_per_second": 69.354,
      "eval_steps_per_second": 3.226,
      "step": 164
    },
    {
      "epoch": 83.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.41790419816970825,
      "eval_runtime": 0.6194,
      "eval_samples_per_second": 69.427,
      "eval_steps_per_second": 3.229,
      "step": 166
    },
    {
      "epoch": 84.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.4044622480869293,
      "eval_runtime": 0.6431,
      "eval_samples_per_second": 66.867,
      "eval_steps_per_second": 3.11,
      "step": 168
    },
    {
      "epoch": 85.0,
      "grad_norm": 1.5876455307006836,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.0634,
      "step": 170
    },
    {
      "epoch": 85.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.4199570417404175,
      "eval_runtime": 0.6161,
      "eval_samples_per_second": 69.796,
      "eval_steps_per_second": 3.246,
      "step": 170
    },
    {
      "epoch": 86.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.4496786296367645,
      "eval_runtime": 0.616,
      "eval_samples_per_second": 69.803,
      "eval_steps_per_second": 3.247,
      "step": 172
    },
    {
      "epoch": 87.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.4796467125415802,
      "eval_runtime": 0.6306,
      "eval_samples_per_second": 68.185,
      "eval_steps_per_second": 3.171,
      "step": 174
    },
    {
      "epoch": 88.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.4996866285800934,
      "eval_runtime": 0.6489,
      "eval_samples_per_second": 66.262,
      "eval_steps_per_second": 3.082,
      "step": 176
    },
    {
      "epoch": 89.0,
      "eval_accuracy": 0.8837209302325582,
      "eval_loss": 0.48668137192726135,
      "eval_runtime": 0.6277,
      "eval_samples_per_second": 68.506,
      "eval_steps_per_second": 3.186,
      "step": 178
    },
    {
      "epoch": 90.0,
      "grad_norm": 2.028071403503418,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.0758,
      "step": 180
    },
    {
      "epoch": 90.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.4478110373020172,
      "eval_runtime": 0.62,
      "eval_samples_per_second": 69.352,
      "eval_steps_per_second": 3.226,
      "step": 180
    },
    {
      "epoch": 91.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.41449663043022156,
      "eval_runtime": 0.6287,
      "eval_samples_per_second": 68.395,
      "eval_steps_per_second": 3.181,
      "step": 182
    },
    {
      "epoch": 92.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.4036337435245514,
      "eval_runtime": 0.6255,
      "eval_samples_per_second": 68.74,
      "eval_steps_per_second": 3.197,
      "step": 184
    },
    {
      "epoch": 93.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.3876798748970032,
      "eval_runtime": 0.6156,
      "eval_samples_per_second": 69.845,
      "eval_steps_per_second": 3.249,
      "step": 186
    },
    {
      "epoch": 94.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.37471088767051697,
      "eval_runtime": 0.6191,
      "eval_samples_per_second": 69.46,
      "eval_steps_per_second": 3.231,
      "step": 188
    },
    {
      "epoch": 95.0,
      "grad_norm": 0.7553315162658691,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.0529,
      "step": 190
    },
    {
      "epoch": 95.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3684982359409332,
      "eval_runtime": 0.6183,
      "eval_samples_per_second": 69.541,
      "eval_steps_per_second": 3.234,
      "step": 190
    },
    {
      "epoch": 96.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.36436358094215393,
      "eval_runtime": 0.62,
      "eval_samples_per_second": 69.354,
      "eval_steps_per_second": 3.226,
      "step": 192
    },
    {
      "epoch": 97.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3648455739021301,
      "eval_runtime": 0.607,
      "eval_samples_per_second": 70.843,
      "eval_steps_per_second": 3.295,
      "step": 194
    },
    {
      "epoch": 98.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.36630958318710327,
      "eval_runtime": 0.6175,
      "eval_samples_per_second": 69.641,
      "eval_steps_per_second": 3.239,
      "step": 196
    },
    {
      "epoch": 99.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3686269223690033,
      "eval_runtime": 0.615,
      "eval_samples_per_second": 69.92,
      "eval_steps_per_second": 3.252,
      "step": 198
    },
    {
      "epoch": 100.0,
      "grad_norm": 1.4840492010116577,
      "learning_rate": 0.0,
      "loss": 0.0594,
      "step": 200
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.9069767441860465,
      "eval_loss": 0.3693346083164215,
      "eval_runtime": 0.6129,
      "eval_samples_per_second": 70.158,
      "eval_steps_per_second": 3.263,
      "step": 200
    },
    {
      "epoch": 100.0,
      "step": 200,
      "total_flos": 1.867590382823424e+18,
      "train_loss": 0.14573614299297333,
      "train_runtime": 1123.1348,
      "train_samples_per_second": 21.458,
      "train_steps_per_second": 0.178
    },
    {
      "epoch": 100.0,
      "eval_accuracy": 0.9302325581395349,
      "eval_loss": 0.25641849637031555,
      "eval_runtime": 0.6659,
      "eval_samples_per_second": 64.579,
      "eval_steps_per_second": 3.004,
      "step": 200
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 100,
  "save_steps": 500,
  "stateful_callbacks": {
    "TrainerControl": {
      "args": {
        "should_epoch_stop": false,
        "should_evaluate": false,
        "should_log": false,
        "should_save": true,
        "should_training_stop": true
      },
      "attributes": {}
    }
  },
  "total_flos": 1.867590382823424e+18,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}