{
  "best_metric": 0.6944444444444444,
  "best_model_checkpoint": "deit-base-distilled-patch16-224-hasta-55-fold3/checkpoint-75",
  "epoch": 57.142857142857146,
  "eval_steps": 500,
  "global_step": 100,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.5714285714285714,
      "eval_accuracy": 0.3611111111111111,
      "eval_loss": 1.3455348014831543,
      "eval_runtime": 0.4812,
      "eval_samples_per_second": 74.811,
      "eval_steps_per_second": 4.156,
      "step": 1
    },
    {
      "epoch": 1.7142857142857144,
      "eval_accuracy": 0.3888888888888889,
      "eval_loss": 1.159979224205017,
      "eval_runtime": 0.5051,
      "eval_samples_per_second": 71.276,
      "eval_steps_per_second": 3.96,
      "step": 3
    },
    {
      "epoch": 2.857142857142857,
      "eval_accuracy": 0.3888888888888889,
      "eval_loss": 1.1451469659805298,
      "eval_runtime": 0.5116,
      "eval_samples_per_second": 70.371,
      "eval_steps_per_second": 3.909,
      "step": 5
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4166666666666667,
      "eval_loss": 1.111316204071045,
      "eval_runtime": 0.5284,
      "eval_samples_per_second": 68.127,
      "eval_steps_per_second": 3.785,
      "step": 7
    },
    {
      "epoch": 4.571428571428571,
      "eval_accuracy": 0.3611111111111111,
      "eval_loss": 1.129367709159851,
      "eval_runtime": 0.503,
      "eval_samples_per_second": 71.577,
      "eval_steps_per_second": 3.977,
      "step": 8
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 2.278644323348999,
      "learning_rate": 5e-05,
      "loss": 1.114,
      "step": 10
    },
    {
      "epoch": 5.714285714285714,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 1.0999935865402222,
      "eval_runtime": 0.5006,
      "eval_samples_per_second": 71.916,
      "eval_steps_per_second": 3.995,
      "step": 10
    },
    {
      "epoch": 6.857142857142857,
      "eval_accuracy": 0.5,
      "eval_loss": 1.0526280403137207,
      "eval_runtime": 0.4893,
      "eval_samples_per_second": 73.568,
      "eval_steps_per_second": 4.087,
      "step": 12
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.0222362279891968,
      "eval_runtime": 0.5113,
      "eval_samples_per_second": 70.407,
      "eval_steps_per_second": 3.911,
      "step": 14
    },
    {
      "epoch": 8.571428571428571,
      "eval_accuracy": 0.4444444444444444,
      "eval_loss": 1.0203337669372559,
      "eval_runtime": 0.4945,
      "eval_samples_per_second": 72.808,
      "eval_steps_per_second": 4.045,
      "step": 15
    },
    {
      "epoch": 9.714285714285714,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.988318681716919,
      "eval_runtime": 0.4952,
      "eval_samples_per_second": 72.698,
      "eval_steps_per_second": 4.039,
      "step": 17
    },
    {
      "epoch": 10.857142857142858,
      "eval_accuracy": 0.5,
      "eval_loss": 0.9685971736907959,
      "eval_runtime": 0.5026,
      "eval_samples_per_second": 71.622,
      "eval_steps_per_second": 3.979,
      "step": 19
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 4.226619243621826,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.928,
      "step": 20
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.9621982574462891,
      "eval_runtime": 0.5167,
      "eval_samples_per_second": 69.674,
      "eval_steps_per_second": 3.871,
      "step": 21
    },
    {
      "epoch": 12.571428571428571,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.9395506381988525,
      "eval_runtime": 0.5053,
      "eval_samples_per_second": 71.245,
      "eval_steps_per_second": 3.958,
      "step": 22
    },
    {
      "epoch": 13.714285714285714,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.967819094657898,
      "eval_runtime": 0.4993,
      "eval_samples_per_second": 72.096,
      "eval_steps_per_second": 4.005,
      "step": 24
    },
    {
      "epoch": 14.857142857142858,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9044739603996277,
      "eval_runtime": 0.5002,
      "eval_samples_per_second": 71.978,
      "eval_steps_per_second": 3.999,
      "step": 26
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.8841844201087952,
      "eval_runtime": 0.4914,
      "eval_samples_per_second": 73.254,
      "eval_steps_per_second": 4.07,
      "step": 28
    },
    {
      "epoch": 16.571428571428573,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.8690497875213623,
      "eval_runtime": 0.5005,
      "eval_samples_per_second": 71.929,
      "eval_steps_per_second": 3.996,
      "step": 29
    },
    {
      "epoch": 17.142857142857142,
      "grad_norm": 5.408860683441162,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.7807,
      "step": 30
    },
    {
      "epoch": 17.714285714285715,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.8539319634437561,
      "eval_runtime": 0.5044,
      "eval_samples_per_second": 71.369,
      "eval_steps_per_second": 3.965,
      "step": 31
    },
    {
      "epoch": 18.857142857142858,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9446122050285339,
      "eval_runtime": 0.4965,
      "eval_samples_per_second": 72.501,
      "eval_steps_per_second": 4.028,
      "step": 33
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.878534734249115,
      "eval_runtime": 0.4979,
      "eval_samples_per_second": 72.309,
      "eval_steps_per_second": 4.017,
      "step": 35
    },
    {
      "epoch": 20.571428571428573,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.8500129580497742,
      "eval_runtime": 0.4975,
      "eval_samples_per_second": 72.361,
      "eval_steps_per_second": 4.02,
      "step": 36
    },
    {
      "epoch": 21.714285714285715,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9317324161529541,
      "eval_runtime": 0.5005,
      "eval_samples_per_second": 71.926,
      "eval_steps_per_second": 3.996,
      "step": 38
    },
    {
      "epoch": 22.857142857142858,
      "grad_norm": 5.6666483879089355,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6419,
      "step": 40
    },
    {
      "epoch": 22.857142857142858,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.9104676842689514,
      "eval_runtime": 0.5248,
      "eval_samples_per_second": 68.595,
      "eval_steps_per_second": 3.811,
      "step": 40
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.95128333568573,
      "eval_runtime": 0.502,
      "eval_samples_per_second": 71.713,
      "eval_steps_per_second": 3.984,
      "step": 42
    },
    {
      "epoch": 24.571428571428573,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.9154585599899292,
      "eval_runtime": 0.5012,
      "eval_samples_per_second": 71.832,
      "eval_steps_per_second": 3.991,
      "step": 43
    },
    {
      "epoch": 25.714285714285715,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.8751881122589111,
      "eval_runtime": 0.4978,
      "eval_samples_per_second": 72.317,
      "eval_steps_per_second": 4.018,
      "step": 45
    },
    {
      "epoch": 26.857142857142858,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.8911417722702026,
      "eval_runtime": 0.5026,
      "eval_samples_per_second": 71.633,
      "eval_steps_per_second": 3.98,
      "step": 47
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.8772248029708862,
      "eval_runtime": 0.503,
      "eval_samples_per_second": 71.57,
      "eval_steps_per_second": 3.976,
      "step": 49
    },
    {
      "epoch": 28.571428571428573,
      "grad_norm": 4.039524078369141,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.5401,
      "step": 50
    },
    {
      "epoch": 28.571428571428573,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.849612832069397,
      "eval_runtime": 0.4949,
      "eval_samples_per_second": 72.741,
      "eval_steps_per_second": 4.041,
      "step": 50
    },
    {
      "epoch": 29.714285714285715,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.8561813235282898,
      "eval_runtime": 0.4959,
      "eval_samples_per_second": 72.593,
      "eval_steps_per_second": 4.033,
      "step": 52
    },
    {
      "epoch": 30.857142857142858,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.83772873878479,
      "eval_runtime": 0.4905,
      "eval_samples_per_second": 73.397,
      "eval_steps_per_second": 4.078,
      "step": 54
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.9968655705451965,
      "eval_runtime": 0.501,
      "eval_samples_per_second": 71.857,
      "eval_steps_per_second": 3.992,
      "step": 56
    },
    {
      "epoch": 32.57142857142857,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 1.0985074043273926,
      "eval_runtime": 0.4932,
      "eval_samples_per_second": 73.0,
      "eval_steps_per_second": 4.056,
      "step": 57
    },
    {
      "epoch": 33.714285714285715,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.9631519317626953,
      "eval_runtime": 0.4974,
      "eval_samples_per_second": 72.381,
      "eval_steps_per_second": 4.021,
      "step": 59
    },
    {
      "epoch": 34.285714285714285,
      "grad_norm": 5.0279154777526855,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.4658,
      "step": 60
    },
    {
      "epoch": 34.857142857142854,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.8651056289672852,
      "eval_runtime": 0.5107,
      "eval_samples_per_second": 70.494,
      "eval_steps_per_second": 3.916,
      "step": 61
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.8730890154838562,
      "eval_runtime": 0.5004,
      "eval_samples_per_second": 71.936,
      "eval_steps_per_second": 3.996,
      "step": 63
    },
    {
      "epoch": 36.57142857142857,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.9147739410400391,
      "eval_runtime": 0.5195,
      "eval_samples_per_second": 69.296,
      "eval_steps_per_second": 3.85,
      "step": 64
    },
    {
      "epoch": 37.714285714285715,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 1.0383161306381226,
      "eval_runtime": 0.5054,
      "eval_samples_per_second": 71.236,
      "eval_steps_per_second": 3.958,
      "step": 66
    },
    {
      "epoch": 38.857142857142854,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.9203394651412964,
      "eval_runtime": 0.4999,
      "eval_samples_per_second": 72.016,
      "eval_steps_per_second": 4.001,
      "step": 68
    },
    {
      "epoch": 40.0,
      "grad_norm": 2.963888645172119,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.3816,
      "step": 70
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.8720214366912842,
      "eval_runtime": 0.4948,
      "eval_samples_per_second": 72.762,
      "eval_steps_per_second": 4.042,
      "step": 70
    },
    {
      "epoch": 40.57142857142857,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.8788578510284424,
      "eval_runtime": 0.4931,
      "eval_samples_per_second": 73.001,
      "eval_steps_per_second": 4.056,
      "step": 71
    },
    {
      "epoch": 41.714285714285715,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.8741800785064697,
      "eval_runtime": 0.4965,
      "eval_samples_per_second": 72.501,
      "eval_steps_per_second": 4.028,
      "step": 73
    },
    {
      "epoch": 42.857142857142854,
      "eval_accuracy": 0.6944444444444444,
      "eval_loss": 0.8865171074867249,
      "eval_runtime": 0.5158,
      "eval_samples_per_second": 69.793,
      "eval_steps_per_second": 3.877,
      "step": 75
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.8931111097335815,
      "eval_runtime": 0.5006,
      "eval_samples_per_second": 71.908,
      "eval_steps_per_second": 3.995,
      "step": 77
    },
    {
      "epoch": 44.57142857142857,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.9036197662353516,
      "eval_runtime": 0.5088,
      "eval_samples_per_second": 70.759,
      "eval_steps_per_second": 3.931,
      "step": 78
    },
    {
      "epoch": 45.714285714285715,
      "grad_norm": 3.1277315616607666,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.337,
      "step": 80
    },
    {
      "epoch": 45.714285714285715,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9182265400886536,
      "eval_runtime": 0.4955,
      "eval_samples_per_second": 72.654,
      "eval_steps_per_second": 4.036,
      "step": 80
    },
    {
      "epoch": 46.857142857142854,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9406058192253113,
      "eval_runtime": 0.4897,
      "eval_samples_per_second": 73.51,
      "eval_steps_per_second": 4.084,
      "step": 82
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.9587271809577942,
      "eval_runtime": 0.4954,
      "eval_samples_per_second": 72.669,
      "eval_steps_per_second": 4.037,
      "step": 84
    },
    {
      "epoch": 48.57142857142857,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9697230458259583,
      "eval_runtime": 0.489,
      "eval_samples_per_second": 73.623,
      "eval_steps_per_second": 4.09,
      "step": 85
    },
    {
      "epoch": 49.714285714285715,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.9818421602249146,
      "eval_runtime": 0.4991,
      "eval_samples_per_second": 72.128,
      "eval_steps_per_second": 4.007,
      "step": 87
    },
    {
      "epoch": 50.857142857142854,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9691570997238159,
      "eval_runtime": 0.4989,
      "eval_samples_per_second": 72.154,
      "eval_steps_per_second": 4.009,
      "step": 89
    },
    {
      "epoch": 51.42857142857143,
      "grad_norm": 3.445671319961548,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.2958,
      "step": 90
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9426124095916748,
      "eval_runtime": 0.5002,
      "eval_samples_per_second": 71.977,
      "eval_steps_per_second": 3.999,
      "step": 91
    },
    {
      "epoch": 52.57142857142857,
      "eval_accuracy": 0.6666666666666666,
      "eval_loss": 0.9373679757118225,
      "eval_runtime": 0.502,
      "eval_samples_per_second": 71.707,
      "eval_steps_per_second": 3.984,
      "step": 92
    },
    {
      "epoch": 53.714285714285715,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9337738752365112,
      "eval_runtime": 0.4971,
      "eval_samples_per_second": 72.42,
      "eval_steps_per_second": 4.023,
      "step": 94
    },
    {
      "epoch": 54.857142857142854,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9336857199668884,
      "eval_runtime": 0.4959,
      "eval_samples_per_second": 72.595,
      "eval_steps_per_second": 4.033,
      "step": 96
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9361027479171753,
      "eval_runtime": 0.5054,
      "eval_samples_per_second": 71.231,
      "eval_steps_per_second": 3.957,
      "step": 98
    },
    {
      "epoch": 56.57142857142857,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9376605153083801,
      "eval_runtime": 0.4938,
      "eval_samples_per_second": 72.898,
      "eval_steps_per_second": 4.05,
      "step": 99
    },
    {
      "epoch": 57.142857142857146,
      "grad_norm": 5.291339874267578,
      "learning_rate": 0.0,
      "loss": 0.2914,
      "step": 100
    },
    {
      "epoch": 57.142857142857146,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.9384960532188416,
      "eval_runtime": 0.4994,
      "eval_samples_per_second": 72.082,
      "eval_steps_per_second": 4.005,
      "step": 100
    },
    {
      "epoch": 57.142857142857146,
      "step": 100,
      "total_flos": 8.770779002957046e+17,
      "train_loss": 0.5776340651512146,
      "train_runtime": 549.4821,
      "train_samples_per_second": 36.034,
      "train_steps_per_second": 0.182
    },
    {
      "epoch": 57.142857142857146,
      "eval_accuracy": 0.6944444444444444,
      "eval_loss": 0.8865171074867249,
      "eval_runtime": 0.5679,
      "eval_samples_per_second": 63.396,
      "eval_steps_per_second": 3.522,
      "step": 100
    }
  ],
  "logging_steps": 10,
  "max_steps": 100,
  "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": 8.770779002957046e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}