{
  "best_metric": 0.039944570511579514,
  "best_model_checkpoint": "./vit-base-beans/checkpoint-100",
  "epoch": 4.0,
  "global_step": 172,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23,
      "learning_rate": 0.00018837209302325584,
      "loss": 0.4782,
      "step": 10
    },
    {
      "epoch": 0.47,
      "learning_rate": 0.00017674418604651164,
      "loss": 0.2147,
      "step": 20
    },
    {
      "epoch": 0.7,
      "learning_rate": 0.00016511627906976747,
      "loss": 0.0922,
      "step": 30
    },
    {
      "epoch": 0.93,
      "learning_rate": 0.00015348837209302327,
      "loss": 0.2727,
      "step": 40
    },
    {
      "epoch": 1.16,
      "learning_rate": 0.0001418604651162791,
      "loss": 0.0407,
      "step": 50
    },
    {
      "epoch": 1.4,
      "learning_rate": 0.0001302325581395349,
      "loss": 0.0933,
      "step": 60
    },
    {
      "epoch": 1.63,
      "learning_rate": 0.00011860465116279071,
      "loss": 0.0547,
      "step": 70
    },
    {
      "epoch": 1.86,
      "learning_rate": 0.00010697674418604651,
      "loss": 0.0502,
      "step": 80
    },
    {
      "epoch": 2.09,
      "learning_rate": 9.534883720930233e-05,
      "loss": 0.0116,
      "step": 90
    },
    {
      "epoch": 2.33,
      "learning_rate": 8.372093023255814e-05,
      "loss": 0.0109,
      "step": 100
    },
    {
      "epoch": 2.33,
      "eval_accuracy": 0.9895287958115183,
      "eval_loss": 0.039944570511579514,
      "eval_runtime": 3.6359,
      "eval_samples_per_second": 52.531,
      "eval_steps_per_second": 6.601,
      "step": 100
    },
    {
      "epoch": 2.56,
      "learning_rate": 7.209302325581396e-05,
      "loss": 0.008,
      "step": 110
    },
    {
      "epoch": 2.79,
      "learning_rate": 6.0465116279069765e-05,
      "loss": 0.0262,
      "step": 120
    },
    {
      "epoch": 3.02,
      "learning_rate": 4.883720930232558e-05,
      "loss": 0.0069,
      "step": 130
    },
    {
      "epoch": 3.26,
      "learning_rate": 3.7209302325581394e-05,
      "loss": 0.0067,
      "step": 140
    },
    {
      "epoch": 3.49,
      "learning_rate": 2.5581395348837212e-05,
      "loss": 0.0061,
      "step": 150
    },
    {
      "epoch": 3.72,
      "learning_rate": 1.3953488372093024e-05,
      "loss": 0.0059,
      "step": 160
    },
    {
      "epoch": 3.95,
      "learning_rate": 2.325581395348837e-06,
      "loss": 0.0057,
      "step": 170
    },
    {
      "epoch": 4.0,
      "step": 172,
      "total_flos": 2.132579554193572e+17,
      "train_loss": 0.08056612370189192,
      "train_runtime": 121.6622,
      "train_samples_per_second": 22.62,
      "train_steps_per_second": 1.414
    }
  ],
  "max_steps": 172,
  "num_train_epochs": 4,
  "total_flos": 2.132579554193572e+17,
  "trial_name": null,
  "trial_params": null
}