{
  "best_metric": 0.14660954475402832,
  "best_model_checkpoint": "./Beit-doc-classification/checkpoint-240",
  "epoch": 18.75,
  "eval_steps": 10,
  "global_step": 300,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.62,
      "learning_rate": 4.8906250000000006e-05,
      "loss": 2.174,
      "step": 10
    },
    {
      "epoch": 0.62,
      "eval_accuracy": 0.6964285714285714,
      "eval_loss": 0.7704058885574341,
      "eval_runtime": 0.2746,
      "eval_samples_per_second": 203.898,
      "eval_steps_per_second": 25.487,
      "step": 10
    },
    {
      "epoch": 1.25,
      "learning_rate": 4.734375e-05,
      "loss": 0.8459,
      "step": 20
    },
    {
      "epoch": 1.25,
      "eval_accuracy": 0.75,
      "eval_loss": 0.6054777503013611,
      "eval_runtime": 0.2697,
      "eval_samples_per_second": 207.624,
      "eval_steps_per_second": 25.953,
      "step": 20
    },
    {
      "epoch": 1.88,
      "learning_rate": 4.5781250000000005e-05,
      "loss": 0.5794,
      "step": 30
    },
    {
      "epoch": 1.88,
      "eval_accuracy": 0.75,
      "eval_loss": 0.4210772216320038,
      "eval_runtime": 0.2721,
      "eval_samples_per_second": 205.818,
      "eval_steps_per_second": 25.727,
      "step": 30
    },
    {
      "epoch": 2.5,
      "learning_rate": 4.421875e-05,
      "loss": 0.356,
      "step": 40
    },
    {
      "epoch": 2.5,
      "eval_accuracy": 0.9285714285714286,
      "eval_loss": 0.23619188368320465,
      "eval_runtime": 0.276,
      "eval_samples_per_second": 202.895,
      "eval_steps_per_second": 25.362,
      "step": 40
    },
    {
      "epoch": 3.12,
      "learning_rate": 4.2656250000000003e-05,
      "loss": 0.2762,
      "step": 50
    },
    {
      "epoch": 3.12,
      "eval_accuracy": 0.9464285714285714,
      "eval_loss": 0.2510121166706085,
      "eval_runtime": 0.2788,
      "eval_samples_per_second": 200.857,
      "eval_steps_per_second": 25.107,
      "step": 50
    },
    {
      "epoch": 3.75,
      "learning_rate": 4.1093750000000006e-05,
      "loss": 0.2331,
      "step": 60
    },
    {
      "epoch": 3.75,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.18232552707195282,
      "eval_runtime": 0.2722,
      "eval_samples_per_second": 205.715,
      "eval_steps_per_second": 25.714,
      "step": 60
    },
    {
      "epoch": 4.38,
      "learning_rate": 3.953125e-05,
      "loss": 0.183,
      "step": 70
    },
    {
      "epoch": 4.38,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.1663789004087448,
      "eval_runtime": 0.2703,
      "eval_samples_per_second": 207.189,
      "eval_steps_per_second": 25.899,
      "step": 70
    },
    {
      "epoch": 5.0,
      "learning_rate": 3.7968750000000005e-05,
      "loss": 0.1171,
      "step": 80
    },
    {
      "epoch": 5.0,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.182296484708786,
      "eval_runtime": 0.2718,
      "eval_samples_per_second": 206.04,
      "eval_steps_per_second": 25.755,
      "step": 80
    },
    {
      "epoch": 5.62,
      "learning_rate": 3.640625e-05,
      "loss": 0.0875,
      "step": 90
    },
    {
      "epoch": 5.62,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.18081991374492645,
      "eval_runtime": 0.2717,
      "eval_samples_per_second": 206.097,
      "eval_steps_per_second": 25.762,
      "step": 90
    },
    {
      "epoch": 6.25,
      "learning_rate": 3.484375e-05,
      "loss": 0.1186,
      "step": 100
    },
    {
      "epoch": 6.25,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.1665651649236679,
      "eval_runtime": 0.2734,
      "eval_samples_per_second": 204.829,
      "eval_steps_per_second": 25.604,
      "step": 100
    },
    {
      "epoch": 6.88,
      "learning_rate": 3.3281250000000006e-05,
      "loss": 0.1174,
      "step": 110
    },
    {
      "epoch": 6.88,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.14654958248138428,
      "eval_runtime": 0.2716,
      "eval_samples_per_second": 206.219,
      "eval_steps_per_second": 25.777,
      "step": 110
    },
    {
      "epoch": 7.5,
      "learning_rate": 3.171875e-05,
      "loss": 0.0519,
      "step": 120
    },
    {
      "epoch": 7.5,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.1532459259033203,
      "eval_runtime": 0.2731,
      "eval_samples_per_second": 205.051,
      "eval_steps_per_second": 25.631,
      "step": 120
    },
    {
      "epoch": 8.12,
      "learning_rate": 3.015625e-05,
      "loss": 0.0338,
      "step": 130
    },
    {
      "epoch": 8.12,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.18999779224395752,
      "eval_runtime": 0.272,
      "eval_samples_per_second": 205.856,
      "eval_steps_per_second": 25.732,
      "step": 130
    },
    {
      "epoch": 8.75,
      "learning_rate": 2.8593750000000004e-05,
      "loss": 0.0241,
      "step": 140
    },
    {
      "epoch": 8.75,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.1611316055059433,
      "eval_runtime": 0.2718,
      "eval_samples_per_second": 206.047,
      "eval_steps_per_second": 25.756,
      "step": 140
    },
    {
      "epoch": 9.38,
      "learning_rate": 2.7031250000000003e-05,
      "loss": 0.0138,
      "step": 150
    },
    {
      "epoch": 9.38,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.17584912478923798,
      "eval_runtime": 0.2713,
      "eval_samples_per_second": 206.411,
      "eval_steps_per_second": 25.801,
      "step": 150
    },
    {
      "epoch": 10.0,
      "learning_rate": 2.5468750000000002e-05,
      "loss": 0.0155,
      "step": 160
    },
    {
      "epoch": 10.0,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.14303383231163025,
      "eval_runtime": 0.2704,
      "eval_samples_per_second": 207.101,
      "eval_steps_per_second": 25.888,
      "step": 160
    },
    {
      "epoch": 10.62,
      "learning_rate": 2.3906250000000002e-05,
      "loss": 0.0478,
      "step": 170
    },
    {
      "epoch": 10.62,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.15157583355903625,
      "eval_runtime": 0.2795,
      "eval_samples_per_second": 200.393,
      "eval_steps_per_second": 25.049,
      "step": 170
    },
    {
      "epoch": 11.25,
      "learning_rate": 2.234375e-05,
      "loss": 0.0228,
      "step": 180
    },
    {
      "epoch": 11.25,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.17853426933288574,
      "eval_runtime": 0.2946,
      "eval_samples_per_second": 190.103,
      "eval_steps_per_second": 23.763,
      "step": 180
    },
    {
      "epoch": 11.88,
      "learning_rate": 2.0781250000000004e-05,
      "loss": 0.0295,
      "step": 190
    },
    {
      "epoch": 11.88,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.16022925078868866,
      "eval_runtime": 0.2758,
      "eval_samples_per_second": 203.034,
      "eval_steps_per_second": 25.379,
      "step": 190
    },
    {
      "epoch": 12.5,
      "learning_rate": 1.9218750000000003e-05,
      "loss": 0.0084,
      "step": 200
    },
    {
      "epoch": 12.5,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.19061991572380066,
      "eval_runtime": 0.281,
      "eval_samples_per_second": 199.27,
      "eval_steps_per_second": 24.909,
      "step": 200
    },
    {
      "epoch": 13.12,
      "learning_rate": 1.7656250000000002e-05,
      "loss": 0.0064,
      "step": 210
    },
    {
      "epoch": 13.12,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.1927066147327423,
      "eval_runtime": 0.272,
      "eval_samples_per_second": 205.909,
      "eval_steps_per_second": 25.739,
      "step": 210
    },
    {
      "epoch": 13.75,
      "learning_rate": 1.609375e-05,
      "loss": 0.0131,
      "step": 220
    },
    {
      "epoch": 13.75,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.17037805914878845,
      "eval_runtime": 0.2756,
      "eval_samples_per_second": 203.19,
      "eval_steps_per_second": 25.399,
      "step": 220
    },
    {
      "epoch": 14.38,
      "learning_rate": 1.4531250000000003e-05,
      "loss": 0.0134,
      "step": 230
    },
    {
      "epoch": 14.38,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.14340999722480774,
      "eval_runtime": 0.2733,
      "eval_samples_per_second": 204.909,
      "eval_steps_per_second": 25.614,
      "step": 230
    },
    {
      "epoch": 15.0,
      "learning_rate": 1.2968750000000002e-05,
      "loss": 0.0082,
      "step": 240
    },
    {
      "epoch": 15.0,
      "eval_accuracy": 0.9821428571428571,
      "eval_loss": 0.14660954475402832,
      "eval_runtime": 0.2748,
      "eval_samples_per_second": 203.761,
      "eval_steps_per_second": 25.47,
      "step": 240
    },
    {
      "epoch": 15.62,
      "learning_rate": 1.140625e-05,
      "loss": 0.0082,
      "step": 250
    },
    {
      "epoch": 15.62,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.15767399966716766,
      "eval_runtime": 0.2783,
      "eval_samples_per_second": 201.226,
      "eval_steps_per_second": 25.153,
      "step": 250
    },
    {
      "epoch": 16.25,
      "learning_rate": 9.84375e-06,
      "loss": 0.0104,
      "step": 260
    },
    {
      "epoch": 16.25,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.17043821513652802,
      "eval_runtime": 0.2775,
      "eval_samples_per_second": 201.786,
      "eval_steps_per_second": 25.223,
      "step": 260
    },
    {
      "epoch": 16.88,
      "learning_rate": 8.28125e-06,
      "loss": 0.0099,
      "step": 270
    },
    {
      "epoch": 16.88,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.1689167469739914,
      "eval_runtime": 0.2808,
      "eval_samples_per_second": 199.425,
      "eval_steps_per_second": 24.928,
      "step": 270
    },
    {
      "epoch": 17.5,
      "learning_rate": 6.71875e-06,
      "loss": 0.0033,
      "step": 280
    },
    {
      "epoch": 17.5,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.1613139808177948,
      "eval_runtime": 0.2796,
      "eval_samples_per_second": 200.293,
      "eval_steps_per_second": 25.037,
      "step": 280
    },
    {
      "epoch": 18.12,
      "learning_rate": 5.15625e-06,
      "loss": 0.0046,
      "step": 290
    },
    {
      "epoch": 18.12,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.15769068896770477,
      "eval_runtime": 0.2801,
      "eval_samples_per_second": 199.909,
      "eval_steps_per_second": 24.989,
      "step": 290
    },
    {
      "epoch": 18.75,
      "learning_rate": 3.5937499999999997e-06,
      "loss": 0.0017,
      "step": 300
    },
    {
      "epoch": 18.75,
      "eval_accuracy": 0.9642857142857143,
      "eval_loss": 0.15903140604496002,
      "eval_runtime": 0.2817,
      "eval_samples_per_second": 198.793,
      "eval_steps_per_second": 24.849,
      "step": 300
    }
  ],
  "logging_steps": 10,
  "max_steps": 320,
  "num_train_epochs": 20,
  "save_steps": 30,
  "total_flos": 7.227949247524454e+17,
  "trial_name": null,
  "trial_params": null
}