trainer_state.json

{
  "best_metric": 0.03401574864983559,
  "best_model_checkpoint": "./output/checkpoint-597",
  "epoch": 3.0,
  "global_step": 597,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.05907629430294037,
      "eval_runtime": 1.3133,
      "eval_samples_per_second": 537.587,
      "eval_steps_per_second": 17.513,
      "step": 199
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.043659381568431854,
      "eval_runtime": 1.3097,
      "eval_samples_per_second": 539.044,
      "eval_steps_per_second": 17.561,
      "step": 398
    },
    {
      "epoch": 2.51,
      "learning_rate": 8.123953098827471e-06,
      "loss": 0.0605,
      "step": 500
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.03401574864983559,
      "eval_runtime": 1.3095,
      "eval_samples_per_second": 539.131,
      "eval_steps_per_second": 17.564,
      "step": 597
    }
  ],
  "max_steps": 597,
  "num_train_epochs": 3,
  "total_flos": 626183508754944.0,
  "trial_name": null,
  "trial_params": null
}