{
  "best_metric": 0.9032258064516129,
  "best_model_checkpoint": "beit-base-patch16-224-OT-alt\\checkpoint-99",
  "epoch": 36.36363636363637,
  "eval_steps": 500,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.91,
      "eval_accuracy": 0.1774193548387097,
      "eval_loss": 1.7093149423599243,
      "eval_runtime": 2.3415,
      "eval_samples_per_second": 26.478,
      "eval_steps_per_second": 1.708,
      "step": 5
    },
    {
      "epoch": 1.82,
      "learning_rate": 2.5e-05,
      "loss": 1.7744,
      "step": 10
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.1774193548387097,
      "eval_loss": 1.6177618503570557,
      "eval_runtime": 2.3771,
      "eval_samples_per_second": 26.083,
      "eval_steps_per_second": 1.683,
      "step": 11
    },
    {
      "epoch": 2.91,
      "eval_accuracy": 0.1774193548387097,
      "eval_loss": 1.473004698753357,
      "eval_runtime": 2.3411,
      "eval_samples_per_second": 26.484,
      "eval_steps_per_second": 1.709,
      "step": 16
    },
    {
      "epoch": 3.64,
      "learning_rate": 5e-05,
      "loss": 1.5823,
      "step": 20
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.1774193548387097,
      "eval_loss": 1.2754011154174805,
      "eval_runtime": 2.3296,
      "eval_samples_per_second": 26.614,
      "eval_steps_per_second": 1.717,
      "step": 22
    },
    {
      "epoch": 4.91,
      "eval_accuracy": 0.5645161290322581,
      "eval_loss": 1.1454869508743286,
      "eval_runtime": 2.374,
      "eval_samples_per_second": 26.117,
      "eval_steps_per_second": 1.685,
      "step": 27
    },
    {
      "epoch": 5.45,
      "learning_rate": 4.722222222222222e-05,
      "loss": 1.27,
      "step": 30
    },
    {
      "epoch": 6.0,
      "eval_accuracy": 0.6290322580645161,
      "eval_loss": 1.0146747827529907,
      "eval_runtime": 2.4966,
      "eval_samples_per_second": 24.834,
      "eval_steps_per_second": 1.602,
      "step": 33
    },
    {
      "epoch": 6.91,
      "eval_accuracy": 0.5483870967741935,
      "eval_loss": 0.9789792895317078,
      "eval_runtime": 2.3346,
      "eval_samples_per_second": 26.557,
      "eval_steps_per_second": 1.713,
      "step": 38
    },
    {
      "epoch": 7.27,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 1.079,
      "step": 40
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.45161290322580644,
      "eval_loss": 1.0473968982696533,
      "eval_runtime": 2.3881,
      "eval_samples_per_second": 25.962,
      "eval_steps_per_second": 1.675,
      "step": 44
    },
    {
      "epoch": 8.91,
      "eval_accuracy": 0.7580645161290323,
      "eval_loss": 0.8796324729919434,
      "eval_runtime": 2.3814,
      "eval_samples_per_second": 26.035,
      "eval_steps_per_second": 1.68,
      "step": 49
    },
    {
      "epoch": 9.09,
      "learning_rate": 4.166666666666667e-05,
      "loss": 1.005,
      "step": 50
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.7741935483870968,
      "eval_loss": 0.7759426832199097,
      "eval_runtime": 2.3246,
      "eval_samples_per_second": 26.672,
      "eval_steps_per_second": 1.721,
      "step": 55
    },
    {
      "epoch": 10.91,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.8479,
      "step": 60
    },
    {
      "epoch": 10.91,
      "eval_accuracy": 0.8225806451612904,
      "eval_loss": 0.7421430349349976,
      "eval_runtime": 2.3181,
      "eval_samples_per_second": 26.747,
      "eval_steps_per_second": 1.726,
      "step": 60
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.8548387096774194,
      "eval_loss": 0.6759869456291199,
      "eval_runtime": 2.3471,
      "eval_samples_per_second": 26.416,
      "eval_steps_per_second": 1.704,
      "step": 66
    },
    {
      "epoch": 12.73,
      "learning_rate": 3.611111111111111e-05,
      "loss": 0.7695,
      "step": 70
    },
    {
      "epoch": 12.91,
      "eval_accuracy": 0.8387096774193549,
      "eval_loss": 0.5932707190513611,
      "eval_runtime": 2.3526,
      "eval_samples_per_second": 26.354,
      "eval_steps_per_second": 1.7,
      "step": 71
    },
    {
      "epoch": 14.0,
      "eval_accuracy": 0.7741935483870968,
      "eval_loss": 0.6372359991073608,
      "eval_runtime": 2.2975,
      "eval_samples_per_second": 26.985,
      "eval_steps_per_second": 1.741,
      "step": 77
    },
    {
      "epoch": 14.55,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6591,
      "step": 80
    },
    {
      "epoch": 14.91,
      "eval_accuracy": 0.8387096774193549,
      "eval_loss": 0.5653398633003235,
      "eval_runtime": 2.2901,
      "eval_samples_per_second": 27.074,
      "eval_steps_per_second": 1.747,
      "step": 82
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.8709677419354839,
      "eval_loss": 0.4950495958328247,
      "eval_runtime": 2.4381,
      "eval_samples_per_second": 25.43,
      "eval_steps_per_second": 1.641,
      "step": 88
    },
    {
      "epoch": 16.36,
      "learning_rate": 3.055555555555556e-05,
      "loss": 0.5675,
      "step": 90
    },
    {
      "epoch": 16.91,
      "eval_accuracy": 0.8225806451612904,
      "eval_loss": 0.5040029287338257,
      "eval_runtime": 2.3556,
      "eval_samples_per_second": 26.321,
      "eval_steps_per_second": 1.698,
      "step": 93
    },
    {
      "epoch": 18.0,
      "eval_accuracy": 0.9032258064516129,
      "eval_loss": 0.4273931682109833,
      "eval_runtime": 2.3296,
      "eval_samples_per_second": 26.614,
      "eval_steps_per_second": 1.717,
      "step": 99
    },
    {
      "epoch": 18.18,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.5134,
      "step": 100
    },
    {
      "epoch": 18.91,
      "eval_accuracy": 0.8548387096774194,
      "eval_loss": 0.461721807718277,
      "eval_runtime": 2.4876,
      "eval_samples_per_second": 24.923,
      "eval_steps_per_second": 1.608,
      "step": 104
    },
    {
      "epoch": 20.0,
      "learning_rate": 2.5e-05,
      "loss": 0.4418,
      "step": 110
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.8870967741935484,
      "eval_loss": 0.4244731366634369,
      "eval_runtime": 2.3501,
      "eval_samples_per_second": 26.382,
      "eval_steps_per_second": 1.702,
      "step": 110
    },
    {
      "epoch": 20.91,
      "eval_accuracy": 0.8387096774193549,
      "eval_loss": 0.4922010600566864,
      "eval_runtime": 2.4186,
      "eval_samples_per_second": 25.635,
      "eval_steps_per_second": 1.654,
      "step": 115
    },
    {
      "epoch": 21.82,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.402,
      "step": 120
    },
    {
      "epoch": 22.0,
      "eval_accuracy": 0.8225806451612904,
      "eval_loss": 0.5112457275390625,
      "eval_runtime": 2.4655,
      "eval_samples_per_second": 25.147,
      "eval_steps_per_second": 1.622,
      "step": 121
    },
    {
      "epoch": 22.91,
      "eval_accuracy": 0.8548387096774194,
      "eval_loss": 0.46961790323257446,
      "eval_runtime": 2.2816,
      "eval_samples_per_second": 27.174,
      "eval_steps_per_second": 1.753,
      "step": 126
    },
    {
      "epoch": 23.64,
      "learning_rate": 1.9444444444444445e-05,
      "loss": 0.4039,
      "step": 130
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.8709677419354839,
      "eval_loss": 0.40138500928878784,
      "eval_runtime": 2.3996,
      "eval_samples_per_second": 25.838,
      "eval_steps_per_second": 1.667,
      "step": 132
    },
    {
      "epoch": 24.91,
      "eval_accuracy": 0.8225806451612904,
      "eval_loss": 0.5005894899368286,
      "eval_runtime": 2.3451,
      "eval_samples_per_second": 26.439,
      "eval_steps_per_second": 1.706,
      "step": 137
    },
    {
      "epoch": 25.45,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.4216,
      "step": 140
    },
    {
      "epoch": 26.0,
      "eval_accuracy": 0.8548387096774194,
      "eval_loss": 0.5351113080978394,
      "eval_runtime": 2.4946,
      "eval_samples_per_second": 24.853,
      "eval_steps_per_second": 1.603,
      "step": 143
    },
    {
      "epoch": 26.91,
      "eval_accuracy": 0.8548387096774194,
      "eval_loss": 0.5202920436859131,
      "eval_runtime": 2.3236,
      "eval_samples_per_second": 26.683,
      "eval_steps_per_second": 1.721,
      "step": 148
    },
    {
      "epoch": 27.27,
      "learning_rate": 1.388888888888889e-05,
      "loss": 0.3593,
      "step": 150
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.8548387096774194,
      "eval_loss": 0.4081813395023346,
      "eval_runtime": 2.3351,
      "eval_samples_per_second": 26.552,
      "eval_steps_per_second": 1.713,
      "step": 154
    },
    {
      "epoch": 28.91,
      "eval_accuracy": 0.8709677419354839,
      "eval_loss": 0.40170377492904663,
      "eval_runtime": 2.3078,
      "eval_samples_per_second": 26.865,
      "eval_steps_per_second": 1.733,
      "step": 159
    },
    {
      "epoch": 29.09,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.3638,
      "step": 160
    },
    {
      "epoch": 30.0,
      "eval_accuracy": 0.8870967741935484,
      "eval_loss": 0.40682506561279297,
      "eval_runtime": 2.5896,
      "eval_samples_per_second": 23.942,
      "eval_steps_per_second": 1.545,
      "step": 165
    },
    {
      "epoch": 30.91,
      "learning_rate": 8.333333333333334e-06,
      "loss": 0.3509,
      "step": 170
    },
    {
      "epoch": 30.91,
      "eval_accuracy": 0.8870967741935484,
      "eval_loss": 0.3990916311740875,
      "eval_runtime": 2.75,
      "eval_samples_per_second": 22.546,
      "eval_steps_per_second": 1.455,
      "step": 170
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.8709677419354839,
      "eval_loss": 0.3964940309524536,
      "eval_runtime": 2.3794,
      "eval_samples_per_second": 26.056,
      "eval_steps_per_second": 1.681,
      "step": 176
    },
    {
      "epoch": 32.73,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.3426,
      "step": 180
    },
    {
      "epoch": 32.91,
      "eval_accuracy": 0.8709677419354839,
      "eval_loss": 0.3921041190624237,
      "eval_runtime": 2.3356,
      "eval_samples_per_second": 26.546,
      "eval_steps_per_second": 1.713,
      "step": 181
    },
    {
      "epoch": 34.0,
      "eval_accuracy": 0.8709677419354839,
      "eval_loss": 0.39981809258461,
      "eval_runtime": 2.4784,
      "eval_samples_per_second": 25.016,
      "eval_steps_per_second": 1.614,
      "step": 187
    },
    {
      "epoch": 34.55,
      "learning_rate": 2.777777777777778e-06,
      "loss": 0.3253,
      "step": 190
    },
    {
      "epoch": 34.91,
      "eval_accuracy": 0.8870967741935484,
      "eval_loss": 0.41015544533729553,
      "eval_runtime": 2.3626,
      "eval_samples_per_second": 26.243,
      "eval_steps_per_second": 1.693,
      "step": 192
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.8870967741935484,
      "eval_loss": 0.4080694913864136,
      "eval_runtime": 2.3858,
      "eval_samples_per_second": 25.987,
      "eval_steps_per_second": 1.677,
      "step": 198
    },
    {
      "epoch": 36.36,
      "learning_rate": 0.0,
      "loss": 0.3085,
      "step": 200
    },
    {
      "epoch": 36.36,
      "eval_accuracy": 0.8870967741935484,
      "eval_loss": 0.40834033489227295,
      "eval_runtime": 2.3711,
      "eval_samples_per_second": 26.149,
      "eval_steps_per_second": 1.687,
      "step": 200
    },
    {
      "epoch": 36.36,
      "step": 200,
      "total_flos": 9.859141601338982e+17,
      "train_loss": 0.6893912732601166,
      "train_runtime": 709.0628,
      "train_samples_per_second": 19.744,
      "train_steps_per_second": 0.282
    }
  ],
  "logging_steps": 10,
  "max_steps": 200,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 40,
  "save_steps": 500,
  "total_flos": 9.859141601338982e+17,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}