{
  "best_metric": 0.6388888888888888,
  "best_model_checkpoint": "deit-base-distilled-patch16-224-hasta-65-fold3/checkpoint-38",
  "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": 0.3888888888888889,
      "eval_loss": 1.0781989097595215,
      "eval_runtime": 0.4988,
      "eval_samples_per_second": 72.179,
      "eval_steps_per_second": 4.01,
      "step": 1
    },
    {
      "epoch": 1.7142857142857144,
      "eval_accuracy": 0.4444444444444444,
      "eval_loss": 1.0919742584228516,
      "eval_runtime": 0.4962,
      "eval_samples_per_second": 72.549,
      "eval_steps_per_second": 4.03,
      "step": 3
    },
    {
      "epoch": 2.857142857142857,
      "eval_accuracy": 0.3333333333333333,
      "eval_loss": 1.1444017887115479,
      "eval_runtime": 0.5029,
      "eval_samples_per_second": 71.59,
      "eval_steps_per_second": 3.977,
      "step": 5
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.4444444444444444,
      "eval_loss": 1.0610096454620361,
      "eval_runtime": 0.5067,
      "eval_samples_per_second": 71.052,
      "eval_steps_per_second": 3.947,
      "step": 7
    },
    {
      "epoch": 4.571428571428571,
      "eval_accuracy": 0.5,
      "eval_loss": 1.0670982599258423,
      "eval_runtime": 0.5087,
      "eval_samples_per_second": 70.762,
      "eval_steps_per_second": 3.931,
      "step": 8
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 5.576918601989746,
      "learning_rate": 5e-05,
      "loss": 1.0571,
      "step": 10
    },
    {
      "epoch": 5.714285714285714,
      "eval_accuracy": 0.4444444444444444,
      "eval_loss": 1.179304599761963,
      "eval_runtime": 0.4963,
      "eval_samples_per_second": 72.537,
      "eval_steps_per_second": 4.03,
      "step": 10
    },
    {
      "epoch": 6.857142857142857,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 1.086612343788147,
      "eval_runtime": 0.487,
      "eval_samples_per_second": 73.923,
      "eval_steps_per_second": 4.107,
      "step": 12
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.1467864513397217,
      "eval_runtime": 0.4832,
      "eval_samples_per_second": 74.51,
      "eval_steps_per_second": 4.139,
      "step": 14
    },
    {
      "epoch": 8.571428571428571,
      "eval_accuracy": 0.4166666666666667,
      "eval_loss": 1.2370481491088867,
      "eval_runtime": 0.4877,
      "eval_samples_per_second": 73.81,
      "eval_steps_per_second": 4.101,
      "step": 15
    },
    {
      "epoch": 9.714285714285714,
      "eval_accuracy": 0.5,
      "eval_loss": 1.0559484958648682,
      "eval_runtime": 0.5002,
      "eval_samples_per_second": 71.967,
      "eval_steps_per_second": 3.998,
      "step": 17
    },
    {
      "epoch": 10.857142857142858,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.9752185344696045,
      "eval_runtime": 0.5095,
      "eval_samples_per_second": 70.655,
      "eval_steps_per_second": 3.925,
      "step": 19
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 8.030916213989258,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 0.8894,
      "step": 20
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.0525615215301514,
      "eval_runtime": 0.4928,
      "eval_samples_per_second": 73.055,
      "eval_steps_per_second": 4.059,
      "step": 21
    },
    {
      "epoch": 12.571428571428571,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 1.0316534042358398,
      "eval_runtime": 0.4988,
      "eval_samples_per_second": 72.176,
      "eval_steps_per_second": 4.01,
      "step": 22
    },
    {
      "epoch": 13.714285714285714,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 1.0261681079864502,
      "eval_runtime": 0.4974,
      "eval_samples_per_second": 72.378,
      "eval_steps_per_second": 4.021,
      "step": 24
    },
    {
      "epoch": 14.857142857142858,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.9768860340118408,
      "eval_runtime": 0.494,
      "eval_samples_per_second": 72.872,
      "eval_steps_per_second": 4.048,
      "step": 26
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.9296702146530151,
      "eval_runtime": 0.4924,
      "eval_samples_per_second": 73.118,
      "eval_steps_per_second": 4.062,
      "step": 28
    },
    {
      "epoch": 16.571428571428573,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.8848314881324768,
      "eval_runtime": 0.4909,
      "eval_samples_per_second": 73.333,
      "eval_steps_per_second": 4.074,
      "step": 29
    },
    {
      "epoch": 17.142857142857142,
      "grad_norm": 4.720308303833008,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.7239,
      "step": 30
    },
    {
      "epoch": 17.714285714285715,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.8514459729194641,
      "eval_runtime": 0.5,
      "eval_samples_per_second": 71.997,
      "eval_steps_per_second": 4.0,
      "step": 31
    },
    {
      "epoch": 18.857142857142858,
      "eval_accuracy": 0.5,
      "eval_loss": 0.8981277942657471,
      "eval_runtime": 0.5083,
      "eval_samples_per_second": 70.828,
      "eval_steps_per_second": 3.935,
      "step": 33
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.5,
      "eval_loss": 0.9670491814613342,
      "eval_runtime": 0.499,
      "eval_samples_per_second": 72.145,
      "eval_steps_per_second": 4.008,
      "step": 35
    },
    {
      "epoch": 20.571428571428573,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.85024094581604,
      "eval_runtime": 0.4914,
      "eval_samples_per_second": 73.257,
      "eval_steps_per_second": 4.07,
      "step": 36
    },
    {
      "epoch": 21.714285714285715,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.7784740328788757,
      "eval_runtime": 0.5025,
      "eval_samples_per_second": 71.636,
      "eval_steps_per_second": 3.98,
      "step": 38
    },
    {
      "epoch": 22.857142857142858,
      "grad_norm": 6.049220561981201,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.6094,
      "step": 40
    },
    {
      "epoch": 22.857142857142858,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.9256177544593811,
      "eval_runtime": 0.4987,
      "eval_samples_per_second": 72.183,
      "eval_steps_per_second": 4.01,
      "step": 40
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.903743326663971,
      "eval_runtime": 0.4838,
      "eval_samples_per_second": 74.409,
      "eval_steps_per_second": 4.134,
      "step": 42
    },
    {
      "epoch": 24.571428571428573,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.8753263354301453,
      "eval_runtime": 0.4952,
      "eval_samples_per_second": 72.701,
      "eval_steps_per_second": 4.039,
      "step": 43
    },
    {
      "epoch": 25.714285714285715,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.8113481402397156,
      "eval_runtime": 0.493,
      "eval_samples_per_second": 73.029,
      "eval_steps_per_second": 4.057,
      "step": 45
    },
    {
      "epoch": 26.857142857142858,
      "eval_accuracy": 0.5,
      "eval_loss": 0.9797316789627075,
      "eval_runtime": 0.4954,
      "eval_samples_per_second": 72.663,
      "eval_steps_per_second": 4.037,
      "step": 47
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.5,
      "eval_loss": 1.0318821668624878,
      "eval_runtime": 0.4986,
      "eval_samples_per_second": 72.196,
      "eval_steps_per_second": 4.011,
      "step": 49
    },
    {
      "epoch": 28.571428571428573,
      "grad_norm": 4.888261318206787,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.4826,
      "step": 50
    },
    {
      "epoch": 28.571428571428573,
      "eval_accuracy": 0.5,
      "eval_loss": 0.9113612771034241,
      "eval_runtime": 0.4876,
      "eval_samples_per_second": 73.834,
      "eval_steps_per_second": 4.102,
      "step": 50
    },
    {
      "epoch": 29.714285714285715,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.7637109160423279,
      "eval_runtime": 0.4915,
      "eval_samples_per_second": 73.252,
      "eval_steps_per_second": 4.07,
      "step": 52
    },
    {
      "epoch": 30.857142857142858,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.80479896068573,
      "eval_runtime": 0.4951,
      "eval_samples_per_second": 72.711,
      "eval_steps_per_second": 4.04,
      "step": 54
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.5,
      "eval_loss": 0.9822417497634888,
      "eval_runtime": 0.4875,
      "eval_samples_per_second": 73.85,
      "eval_steps_per_second": 4.103,
      "step": 56
    },
    {
      "epoch": 32.57142857142857,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.9030971527099609,
      "eval_runtime": 0.4931,
      "eval_samples_per_second": 73.011,
      "eval_steps_per_second": 4.056,
      "step": 57
    },
    {
      "epoch": 33.714285714285715,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.7210705280303955,
      "eval_runtime": 0.4818,
      "eval_samples_per_second": 74.719,
      "eval_steps_per_second": 4.151,
      "step": 59
    },
    {
      "epoch": 34.285714285714285,
      "grad_norm": 5.53534460067749,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.3943,
      "step": 60
    },
    {
      "epoch": 34.857142857142854,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.6978909969329834,
      "eval_runtime": 0.4896,
      "eval_samples_per_second": 73.536,
      "eval_steps_per_second": 4.085,
      "step": 61
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.7323928475379944,
      "eval_runtime": 0.4914,
      "eval_samples_per_second": 73.26,
      "eval_steps_per_second": 4.07,
      "step": 63
    },
    {
      "epoch": 36.57142857142857,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.7461583018302917,
      "eval_runtime": 0.5029,
      "eval_samples_per_second": 71.586,
      "eval_steps_per_second": 3.977,
      "step": 64
    },
    {
      "epoch": 37.714285714285715,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.7727600932121277,
      "eval_runtime": 0.4906,
      "eval_samples_per_second": 73.38,
      "eval_steps_per_second": 4.077,
      "step": 66
    },
    {
      "epoch": 38.857142857142854,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.753002941608429,
      "eval_runtime": 0.4949,
      "eval_samples_per_second": 72.747,
      "eval_steps_per_second": 4.041,
      "step": 68
    },
    {
      "epoch": 40.0,
      "grad_norm": 5.077225208282471,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.3325,
      "step": 70
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.7361128926277161,
      "eval_runtime": 0.4916,
      "eval_samples_per_second": 73.235,
      "eval_steps_per_second": 4.069,
      "step": 70
    },
    {
      "epoch": 40.57142857142857,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.7226569652557373,
      "eval_runtime": 0.4863,
      "eval_samples_per_second": 74.034,
      "eval_steps_per_second": 4.113,
      "step": 71
    },
    {
      "epoch": 41.714285714285715,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.7938054800033569,
      "eval_runtime": 0.4921,
      "eval_samples_per_second": 73.158,
      "eval_steps_per_second": 4.064,
      "step": 73
    },
    {
      "epoch": 42.857142857142854,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.8002898693084717,
      "eval_runtime": 0.4883,
      "eval_samples_per_second": 73.73,
      "eval_steps_per_second": 4.096,
      "step": 75
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.754361093044281,
      "eval_runtime": 0.4969,
      "eval_samples_per_second": 72.444,
      "eval_steps_per_second": 4.025,
      "step": 77
    },
    {
      "epoch": 44.57142857142857,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.7556236982345581,
      "eval_runtime": 0.4969,
      "eval_samples_per_second": 72.45,
      "eval_steps_per_second": 4.025,
      "step": 78
    },
    {
      "epoch": 45.714285714285715,
      "grad_norm": 3.50972843170166,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.2911,
      "step": 80
    },
    {
      "epoch": 45.714285714285715,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.7857760190963745,
      "eval_runtime": 0.5038,
      "eval_samples_per_second": 71.451,
      "eval_steps_per_second": 3.97,
      "step": 80
    },
    {
      "epoch": 46.857142857142854,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.799206018447876,
      "eval_runtime": 0.5005,
      "eval_samples_per_second": 71.93,
      "eval_steps_per_second": 3.996,
      "step": 82
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.8293063640594482,
      "eval_runtime": 0.4894,
      "eval_samples_per_second": 73.56,
      "eval_steps_per_second": 4.087,
      "step": 84
    },
    {
      "epoch": 48.57142857142857,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.8294222354888916,
      "eval_runtime": 0.488,
      "eval_samples_per_second": 73.778,
      "eval_steps_per_second": 4.099,
      "step": 85
    },
    {
      "epoch": 49.714285714285715,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.8113357424736023,
      "eval_runtime": 0.4924,
      "eval_samples_per_second": 73.105,
      "eval_steps_per_second": 4.061,
      "step": 87
    },
    {
      "epoch": 50.857142857142854,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.8061643242835999,
      "eval_runtime": 0.5103,
      "eval_samples_per_second": 70.542,
      "eval_steps_per_second": 3.919,
      "step": 89
    },
    {
      "epoch": 51.42857142857143,
      "grad_norm": 4.2429914474487305,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.2577,
      "step": 90
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.8508003950119019,
      "eval_runtime": 0.4998,
      "eval_samples_per_second": 72.032,
      "eval_steps_per_second": 4.002,
      "step": 91
    },
    {
      "epoch": 52.57142857142857,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.874424397945404,
      "eval_runtime": 0.5048,
      "eval_samples_per_second": 71.314,
      "eval_steps_per_second": 3.962,
      "step": 92
    },
    {
      "epoch": 53.714285714285715,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.894795298576355,
      "eval_runtime": 0.4866,
      "eval_samples_per_second": 73.986,
      "eval_steps_per_second": 4.11,
      "step": 94
    },
    {
      "epoch": 54.857142857142854,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.8976125717163086,
      "eval_runtime": 0.4871,
      "eval_samples_per_second": 73.913,
      "eval_steps_per_second": 4.106,
      "step": 96
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.8932664394378662,
      "eval_runtime": 0.5026,
      "eval_samples_per_second": 71.622,
      "eval_steps_per_second": 3.979,
      "step": 98
    },
    {
      "epoch": 56.57142857142857,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.8899792432785034,
      "eval_runtime": 0.4931,
      "eval_samples_per_second": 73.013,
      "eval_steps_per_second": 4.056,
      "step": 99
    },
    {
      "epoch": 57.142857142857146,
      "grad_norm": 3.8109960556030273,
      "learning_rate": 0.0,
      "loss": 0.2642,
      "step": 100
    },
    {
      "epoch": 57.142857142857146,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.8885748386383057,
      "eval_runtime": 0.4937,
      "eval_samples_per_second": 72.912,
      "eval_steps_per_second": 4.051,
      "step": 100
    },
    {
      "epoch": 57.142857142857146,
      "step": 100,
      "total_flos": 8.903293865079828e+17,
      "train_loss": 0.5302144145965576,
      "train_runtime": 562.7626,
      "train_samples_per_second": 35.717,
      "train_steps_per_second": 0.178
    },
    {
      "epoch": 57.142857142857146,
      "eval_accuracy": 0.6388888888888888,
      "eval_loss": 0.7784740328788757,
      "eval_runtime": 0.5378,
      "eval_samples_per_second": 66.935,
      "eval_steps_per_second": 3.719,
      "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": 8.903293865079828e+17,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": null
}