{
  "best_metric": 0.6111111111111112,
  "best_model_checkpoint": "deit-base-distilled-patch16-224-hasta-65-fold2/checkpoint-68",
  "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": 1.2209676504135132,
      "eval_runtime": 0.4959,
      "eval_samples_per_second": 72.601,
      "eval_steps_per_second": 4.033,
      "step": 1
    },
    {
      "epoch": 1.7142857142857144,
      "eval_accuracy": 0.2777777777777778,
      "eval_loss": 1.184085488319397,
      "eval_runtime": 0.502,
      "eval_samples_per_second": 71.713,
      "eval_steps_per_second": 3.984,
      "step": 3
    },
    {
      "epoch": 2.857142857142857,
      "eval_accuracy": 0.2777777777777778,
      "eval_loss": 1.3488738536834717,
      "eval_runtime": 0.5074,
      "eval_samples_per_second": 70.954,
      "eval_steps_per_second": 3.942,
      "step": 5
    },
    {
      "epoch": 4.0,
      "eval_accuracy": 0.2777777777777778,
      "eval_loss": 1.2177715301513672,
      "eval_runtime": 0.518,
      "eval_samples_per_second": 69.499,
      "eval_steps_per_second": 3.861,
      "step": 7
    },
    {
      "epoch": 4.571428571428571,
      "eval_accuracy": 0.2222222222222222,
      "eval_loss": 1.1297016143798828,
      "eval_runtime": 0.4898,
      "eval_samples_per_second": 73.502,
      "eval_steps_per_second": 4.083,
      "step": 8
    },
    {
      "epoch": 5.714285714285714,
      "grad_norm": 2.9522297382354736,
      "learning_rate": 5e-05,
      "loss": 1.1666,
      "step": 10
    },
    {
      "epoch": 5.714285714285714,
      "eval_accuracy": 0.3055555555555556,
      "eval_loss": 1.12113356590271,
      "eval_runtime": 0.4846,
      "eval_samples_per_second": 74.289,
      "eval_steps_per_second": 4.127,
      "step": 10
    },
    {
      "epoch": 6.857142857142857,
      "eval_accuracy": 0.4166666666666667,
      "eval_loss": 1.0956040620803833,
      "eval_runtime": 0.4826,
      "eval_samples_per_second": 74.6,
      "eval_steps_per_second": 4.144,
      "step": 12
    },
    {
      "epoch": 8.0,
      "eval_accuracy": 0.3055555555555556,
      "eval_loss": 1.0999130010604858,
      "eval_runtime": 0.4861,
      "eval_samples_per_second": 74.052,
      "eval_steps_per_second": 4.114,
      "step": 14
    },
    {
      "epoch": 8.571428571428571,
      "eval_accuracy": 0.4166666666666667,
      "eval_loss": 1.103540062904358,
      "eval_runtime": 0.4943,
      "eval_samples_per_second": 72.83,
      "eval_steps_per_second": 4.046,
      "step": 15
    },
    {
      "epoch": 9.714285714285714,
      "eval_accuracy": 0.4166666666666667,
      "eval_loss": 1.0611950159072876,
      "eval_runtime": 0.502,
      "eval_samples_per_second": 71.715,
      "eval_steps_per_second": 3.984,
      "step": 17
    },
    {
      "epoch": 10.857142857142858,
      "eval_accuracy": 0.5,
      "eval_loss": 1.0405381917953491,
      "eval_runtime": 0.4952,
      "eval_samples_per_second": 72.692,
      "eval_steps_per_second": 4.038,
      "step": 19
    },
    {
      "epoch": 11.428571428571429,
      "grad_norm": 4.649017810821533,
      "learning_rate": 4.4444444444444447e-05,
      "loss": 1.0161,
      "step": 20
    },
    {
      "epoch": 12.0,
      "eval_accuracy": 0.3888888888888889,
      "eval_loss": 1.0977853536605835,
      "eval_runtime": 0.4898,
      "eval_samples_per_second": 73.504,
      "eval_steps_per_second": 4.084,
      "step": 21
    },
    {
      "epoch": 12.571428571428571,
      "eval_accuracy": 0.3888888888888889,
      "eval_loss": 1.010968804359436,
      "eval_runtime": 0.4985,
      "eval_samples_per_second": 72.223,
      "eval_steps_per_second": 4.012,
      "step": 22
    },
    {
      "epoch": 13.714285714285714,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 1.0061941146850586,
      "eval_runtime": 0.502,
      "eval_samples_per_second": 71.712,
      "eval_steps_per_second": 3.984,
      "step": 24
    },
    {
      "epoch": 14.857142857142858,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.9770500659942627,
      "eval_runtime": 0.4917,
      "eval_samples_per_second": 73.223,
      "eval_steps_per_second": 4.068,
      "step": 26
    },
    {
      "epoch": 16.0,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.9987720847129822,
      "eval_runtime": 0.4962,
      "eval_samples_per_second": 72.55,
      "eval_steps_per_second": 4.031,
      "step": 28
    },
    {
      "epoch": 16.571428571428573,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.9966822862625122,
      "eval_runtime": 0.496,
      "eval_samples_per_second": 72.575,
      "eval_steps_per_second": 4.032,
      "step": 29
    },
    {
      "epoch": 17.142857142857142,
      "grad_norm": 5.285099983215332,
      "learning_rate": 3.888888888888889e-05,
      "loss": 0.9177,
      "step": 30
    },
    {
      "epoch": 17.714285714285715,
      "eval_accuracy": 0.4444444444444444,
      "eval_loss": 0.999761164188385,
      "eval_runtime": 0.5028,
      "eval_samples_per_second": 71.597,
      "eval_steps_per_second": 3.978,
      "step": 31
    },
    {
      "epoch": 18.857142857142858,
      "eval_accuracy": 0.5,
      "eval_loss": 1.0774204730987549,
      "eval_runtime": 0.4961,
      "eval_samples_per_second": 72.566,
      "eval_steps_per_second": 4.031,
      "step": 33
    },
    {
      "epoch": 20.0,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.9775403141975403,
      "eval_runtime": 0.4953,
      "eval_samples_per_second": 72.679,
      "eval_steps_per_second": 4.038,
      "step": 35
    },
    {
      "epoch": 20.571428571428573,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.9917855858802795,
      "eval_runtime": 0.4929,
      "eval_samples_per_second": 73.035,
      "eval_steps_per_second": 4.058,
      "step": 36
    },
    {
      "epoch": 21.714285714285715,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 1.0066277980804443,
      "eval_runtime": 0.4973,
      "eval_samples_per_second": 72.393,
      "eval_steps_per_second": 4.022,
      "step": 38
    },
    {
      "epoch": 22.857142857142858,
      "grad_norm": 9.40731430053711,
      "learning_rate": 3.3333333333333335e-05,
      "loss": 0.7319,
      "step": 40
    },
    {
      "epoch": 22.857142857142858,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 1.0558935403823853,
      "eval_runtime": 0.4948,
      "eval_samples_per_second": 72.754,
      "eval_steps_per_second": 4.042,
      "step": 40
    },
    {
      "epoch": 24.0,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 1.0745033025741577,
      "eval_runtime": 0.5016,
      "eval_samples_per_second": 71.773,
      "eval_steps_per_second": 3.987,
      "step": 42
    },
    {
      "epoch": 24.571428571428573,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 1.0611257553100586,
      "eval_runtime": 0.4907,
      "eval_samples_per_second": 73.364,
      "eval_steps_per_second": 4.076,
      "step": 43
    },
    {
      "epoch": 25.714285714285715,
      "eval_accuracy": 0.4444444444444444,
      "eval_loss": 0.9831328988075256,
      "eval_runtime": 0.4964,
      "eval_samples_per_second": 72.516,
      "eval_steps_per_second": 4.029,
      "step": 45
    },
    {
      "epoch": 26.857142857142858,
      "eval_accuracy": 0.4444444444444444,
      "eval_loss": 1.035706877708435,
      "eval_runtime": 0.5065,
      "eval_samples_per_second": 71.082,
      "eval_steps_per_second": 3.949,
      "step": 47
    },
    {
      "epoch": 28.0,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 1.1500649452209473,
      "eval_runtime": 0.4901,
      "eval_samples_per_second": 73.448,
      "eval_steps_per_second": 4.08,
      "step": 49
    },
    {
      "epoch": 28.571428571428573,
      "grad_norm": 8.14036750793457,
      "learning_rate": 2.777777777777778e-05,
      "loss": 0.6173,
      "step": 50
    },
    {
      "epoch": 28.571428571428573,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 1.1570887565612793,
      "eval_runtime": 0.5018,
      "eval_samples_per_second": 71.741,
      "eval_steps_per_second": 3.986,
      "step": 50
    },
    {
      "epoch": 29.714285714285715,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.9706482291221619,
      "eval_runtime": 0.4949,
      "eval_samples_per_second": 72.741,
      "eval_steps_per_second": 4.041,
      "step": 52
    },
    {
      "epoch": 30.857142857142858,
      "eval_accuracy": 0.4444444444444444,
      "eval_loss": 1.0835765600204468,
      "eval_runtime": 0.4925,
      "eval_samples_per_second": 73.098,
      "eval_steps_per_second": 4.061,
      "step": 54
    },
    {
      "epoch": 32.0,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.9925559163093567,
      "eval_runtime": 0.4903,
      "eval_samples_per_second": 73.419,
      "eval_steps_per_second": 4.079,
      "step": 56
    },
    {
      "epoch": 32.57142857142857,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.9647738337516785,
      "eval_runtime": 0.4909,
      "eval_samples_per_second": 73.329,
      "eval_steps_per_second": 4.074,
      "step": 57
    },
    {
      "epoch": 33.714285714285715,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 1.0512733459472656,
      "eval_runtime": 0.4952,
      "eval_samples_per_second": 72.699,
      "eval_steps_per_second": 4.039,
      "step": 59
    },
    {
      "epoch": 34.285714285714285,
      "grad_norm": 15.036521911621094,
      "learning_rate": 2.2222222222222223e-05,
      "loss": 0.5518,
      "step": 60
    },
    {
      "epoch": 34.857142857142854,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.9229605197906494,
      "eval_runtime": 0.5131,
      "eval_samples_per_second": 70.168,
      "eval_steps_per_second": 3.898,
      "step": 61
    },
    {
      "epoch": 36.0,
      "eval_accuracy": 0.4444444444444444,
      "eval_loss": 0.949390709400177,
      "eval_runtime": 0.4893,
      "eval_samples_per_second": 73.574,
      "eval_steps_per_second": 4.087,
      "step": 63
    },
    {
      "epoch": 36.57142857142857,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.9940825700759888,
      "eval_runtime": 0.4868,
      "eval_samples_per_second": 73.955,
      "eval_steps_per_second": 4.109,
      "step": 64
    },
    {
      "epoch": 37.714285714285715,
      "eval_accuracy": 0.5,
      "eval_loss": 0.9322630763053894,
      "eval_runtime": 0.4863,
      "eval_samples_per_second": 74.033,
      "eval_steps_per_second": 4.113,
      "step": 66
    },
    {
      "epoch": 38.857142857142854,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.8775839805603027,
      "eval_runtime": 0.4907,
      "eval_samples_per_second": 73.367,
      "eval_steps_per_second": 4.076,
      "step": 68
    },
    {
      "epoch": 40.0,
      "grad_norm": 5.310029983520508,
      "learning_rate": 1.6666666666666667e-05,
      "loss": 0.512,
      "step": 70
    },
    {
      "epoch": 40.0,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.9269329905509949,
      "eval_runtime": 0.496,
      "eval_samples_per_second": 72.574,
      "eval_steps_per_second": 4.032,
      "step": 70
    },
    {
      "epoch": 40.57142857142857,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.9188339710235596,
      "eval_runtime": 0.5037,
      "eval_samples_per_second": 71.469,
      "eval_steps_per_second": 3.971,
      "step": 71
    },
    {
      "epoch": 41.714285714285715,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.9325751066207886,
      "eval_runtime": 0.4925,
      "eval_samples_per_second": 73.103,
      "eval_steps_per_second": 4.061,
      "step": 73
    },
    {
      "epoch": 42.857142857142854,
      "eval_accuracy": 0.5,
      "eval_loss": 0.9403908848762512,
      "eval_runtime": 0.4988,
      "eval_samples_per_second": 72.177,
      "eval_steps_per_second": 4.01,
      "step": 75
    },
    {
      "epoch": 44.0,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.904733419418335,
      "eval_runtime": 0.4959,
      "eval_samples_per_second": 72.6,
      "eval_steps_per_second": 4.033,
      "step": 77
    },
    {
      "epoch": 44.57142857142857,
      "eval_accuracy": 0.5277777777777778,
      "eval_loss": 0.894720196723938,
      "eval_runtime": 0.4916,
      "eval_samples_per_second": 73.232,
      "eval_steps_per_second": 4.068,
      "step": 78
    },
    {
      "epoch": 45.714285714285715,
      "grad_norm": 5.139469146728516,
      "learning_rate": 1.1111111111111112e-05,
      "loss": 0.4374,
      "step": 80
    },
    {
      "epoch": 45.714285714285715,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.896452784538269,
      "eval_runtime": 0.4882,
      "eval_samples_per_second": 73.743,
      "eval_steps_per_second": 4.097,
      "step": 80
    },
    {
      "epoch": 46.857142857142854,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.9076719284057617,
      "eval_runtime": 0.4989,
      "eval_samples_per_second": 72.154,
      "eval_steps_per_second": 4.009,
      "step": 82
    },
    {
      "epoch": 48.0,
      "eval_accuracy": 0.5,
      "eval_loss": 0.9289677739143372,
      "eval_runtime": 0.4975,
      "eval_samples_per_second": 72.366,
      "eval_steps_per_second": 4.02,
      "step": 84
    },
    {
      "epoch": 48.57142857142857,
      "eval_accuracy": 0.5,
      "eval_loss": 0.9194160103797913,
      "eval_runtime": 0.4986,
      "eval_samples_per_second": 72.207,
      "eval_steps_per_second": 4.011,
      "step": 85
    },
    {
      "epoch": 49.714285714285715,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.8922841548919678,
      "eval_runtime": 0.4863,
      "eval_samples_per_second": 74.03,
      "eval_steps_per_second": 4.113,
      "step": 87
    },
    {
      "epoch": 50.857142857142854,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.8754000067710876,
      "eval_runtime": 0.4986,
      "eval_samples_per_second": 72.206,
      "eval_steps_per_second": 4.011,
      "step": 89
    },
    {
      "epoch": 51.42857142857143,
      "grad_norm": 4.2079758644104,
      "learning_rate": 5.555555555555556e-06,
      "loss": 0.3571,
      "step": 90
    },
    {
      "epoch": 52.0,
      "eval_accuracy": 0.5833333333333334,
      "eval_loss": 0.8766882419586182,
      "eval_runtime": 0.4939,
      "eval_samples_per_second": 72.89,
      "eval_steps_per_second": 4.049,
      "step": 91
    },
    {
      "epoch": 52.57142857142857,
      "eval_accuracy": 0.5555555555555556,
      "eval_loss": 0.8808281421661377,
      "eval_runtime": 0.4979,
      "eval_samples_per_second": 72.301,
      "eval_steps_per_second": 4.017,
      "step": 92
    },
    {
      "epoch": 53.714285714285715,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.8938834071159363,
      "eval_runtime": 0.4947,
      "eval_samples_per_second": 72.776,
      "eval_steps_per_second": 4.043,
      "step": 94
    },
    {
      "epoch": 54.857142857142854,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.9078287482261658,
      "eval_runtime": 0.5093,
      "eval_samples_per_second": 70.681,
      "eval_steps_per_second": 3.927,
      "step": 96
    },
    {
      "epoch": 56.0,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.916972815990448,
      "eval_runtime": 0.5017,
      "eval_samples_per_second": 71.752,
      "eval_steps_per_second": 3.986,
      "step": 98
    },
    {
      "epoch": 56.57142857142857,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.9171713590621948,
      "eval_runtime": 0.4982,
      "eval_samples_per_second": 72.258,
      "eval_steps_per_second": 4.014,
      "step": 99
    },
    {
      "epoch": 57.142857142857146,
      "grad_norm": 3.1942052841186523,
      "learning_rate": 0.0,
      "loss": 0.3333,
      "step": 100
    },
    {
      "epoch": 57.142857142857146,
      "eval_accuracy": 0.4722222222222222,
      "eval_loss": 0.9167850017547607,
      "eval_runtime": 0.4887,
      "eval_samples_per_second": 73.67,
      "eval_steps_per_second": 4.093,
      "step": 100
    },
    {
      "epoch": 57.142857142857146,
      "step": 100,
      "total_flos": 8.903293865079828e+17,
      "train_loss": 0.6641335129737854,
      "train_runtime": 555.1092,
      "train_samples_per_second": 36.209,
      "train_steps_per_second": 0.18
    },
    {
      "epoch": 57.142857142857146,
      "eval_accuracy": 0.6111111111111112,
      "eval_loss": 0.8775839805603027,
      "eval_runtime": 0.5296,
      "eval_samples_per_second": 67.98,
      "eval_steps_per_second": 3.777,
      "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
}