trainer_state.json

{
  "best_metric": 0.09070728719234467,
  "best_model_checkpoint": "./vit-base-beans/checkpoint-200",
  "epoch": 3.571428571428571,
  "global_step": 200,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.18,
      "learning_rate": 0.00019107142857142858,
      "loss": 1.6062,
      "step": 10
    },
    {
      "epoch": 0.36,
      "learning_rate": 0.00018214285714285714,
      "loss": 1.0317,
      "step": 20
    },
    {
      "epoch": 0.54,
      "learning_rate": 0.00017321428571428574,
      "loss": 0.7148,
      "step": 30
    },
    {
      "epoch": 0.71,
      "learning_rate": 0.00016428571428571428,
      "loss": 0.4595,
      "step": 40
    },
    {
      "epoch": 0.89,
      "learning_rate": 0.00015535714285714287,
      "loss": 0.3494,
      "step": 50
    },
    {
      "epoch": 1.07,
      "learning_rate": 0.00014642857142857141,
      "loss": 0.2257,
      "step": 60
    },
    {
      "epoch": 1.25,
      "learning_rate": 0.0001375,
      "loss": 0.1513,
      "step": 70
    },
    {
      "epoch": 1.43,
      "learning_rate": 0.00012857142857142858,
      "loss": 0.1339,
      "step": 80
    },
    {
      "epoch": 1.61,
      "learning_rate": 0.00011964285714285714,
      "loss": 0.1033,
      "step": 90
    },
    {
      "epoch": 1.79,
      "learning_rate": 0.00011071428571428572,
      "loss": 0.1085,
      "step": 100
    },
    {
      "epoch": 1.79,
      "eval_accuracy": 0.967479674796748,
      "eval_loss": 0.15753066539764404,
      "eval_runtime": 3.8142,
      "eval_samples_per_second": 64.496,
      "eval_steps_per_second": 8.128,
      "step": 100
    },
    {
      "epoch": 1.96,
      "learning_rate": 0.00010178571428571428,
      "loss": 0.1405,
      "step": 110
    },
    {
      "epoch": 2.14,
      "learning_rate": 9.285714285714286e-05,
      "loss": 0.0774,
      "step": 120
    },
    {
      "epoch": 2.32,
      "learning_rate": 8.392857142857144e-05,
      "loss": 0.0794,
      "step": 130
    },
    {
      "epoch": 2.5,
      "learning_rate": 7.500000000000001e-05,
      "loss": 0.0594,
      "step": 140
    },
    {
      "epoch": 2.68,
      "learning_rate": 6.607142857142857e-05,
      "loss": 0.0563,
      "step": 150
    },
    {
      "epoch": 2.86,
      "learning_rate": 5.714285714285714e-05,
      "loss": 0.0509,
      "step": 160
    },
    {
      "epoch": 3.04,
      "learning_rate": 4.8214285714285716e-05,
      "loss": 0.0511,
      "step": 170
    },
    {
      "epoch": 3.21,
      "learning_rate": 3.928571428571429e-05,
      "loss": 0.0473,
      "step": 180
    },
    {
      "epoch": 3.39,
      "learning_rate": 3.0357142857142857e-05,
      "loss": 0.0438,
      "step": 190
    },
    {
      "epoch": 3.57,
      "learning_rate": 2.1428571428571428e-05,
      "loss": 0.0429,
      "step": 200
    },
    {
      "epoch": 3.57,
      "eval_accuracy": 0.9878048780487805,
      "eval_loss": 0.09070728719234467,
      "eval_runtime": 4.0141,
      "eval_samples_per_second": 61.283,
      "eval_steps_per_second": 7.723,
      "step": 200
    }
  ],
  "max_steps": 224,
  "num_train_epochs": 4,
  "total_flos": 2.4519564270312653e+17,
  "trial_name": null,
  "trial_params": null
}