{
  "best_metric": 0.26001426577568054,
  "best_model_checkpoint": "./vit-base-beans/checkpoint-120",
  "epoch": 3.1578947368421053,
  "global_step": 120,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.26,
      "learning_rate": 0.00018684210526315792,
      "loss": 0.8842,
      "step": 10
    },
    {
      "epoch": 0.53,
      "learning_rate": 0.0001736842105263158,
      "loss": 0.4388,
      "step": 20
    },
    {
      "epoch": 0.79,
      "learning_rate": 0.0001605263157894737,
      "loss": 0.3165,
      "step": 30
    },
    {
      "epoch": 1.05,
      "learning_rate": 0.00014736842105263158,
      "loss": 0.1621,
      "step": 40
    },
    {
      "epoch": 1.05,
      "eval_accuracy": 0.8820224719101124,
      "eval_loss": 0.3338429927825928,
      "eval_runtime": 3.9187,
      "eval_samples_per_second": 45.423,
      "eval_steps_per_second": 5.869,
      "step": 40
    },
    {
      "epoch": 1.32,
      "learning_rate": 0.00013421052631578948,
      "loss": 0.0843,
      "step": 50
    },
    {
      "epoch": 1.58,
      "learning_rate": 0.00012105263157894738,
      "loss": 0.1241,
      "step": 60
    },
    {
      "epoch": 1.84,
      "learning_rate": 0.00010789473684210527,
      "loss": 0.0418,
      "step": 70
    },
    {
      "epoch": 2.11,
      "learning_rate": 9.473684210526316e-05,
      "loss": 0.0497,
      "step": 80
    },
    {
      "epoch": 2.11,
      "eval_accuracy": 0.8876404494382022,
      "eval_loss": 0.31701481342315674,
      "eval_runtime": 5.1269,
      "eval_samples_per_second": 34.719,
      "eval_steps_per_second": 4.486,
      "step": 80
    },
    {
      "epoch": 2.37,
      "learning_rate": 8.157894736842105e-05,
      "loss": 0.057,
      "step": 90
    },
    {
      "epoch": 2.63,
      "learning_rate": 6.842105263157895e-05,
      "loss": 0.0389,
      "step": 100
    },
    {
      "epoch": 2.89,
      "learning_rate": 5.526315789473685e-05,
      "loss": 0.0211,
      "step": 110
    },
    {
      "epoch": 3.16,
      "learning_rate": 4.210526315789474e-05,
      "loss": 0.0357,
      "step": 120
    },
    {
      "epoch": 3.16,
      "eval_accuracy": 0.9157303370786517,
      "eval_loss": 0.26001426577568054,
      "eval_runtime": 4.1994,
      "eval_samples_per_second": 42.387,
      "eval_steps_per_second": 5.477,
      "step": 120
    }
  ],
  "max_steps": 152,
  "num_train_epochs": 4,
  "total_flos": 1.4692612914988646e+17,
  "trial_name": null,
  "trial_params": null
}