trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 561,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.89,
      "learning_rate": 5.436720142602496e-06,
      "loss": 1.3247,
      "step": 500
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.45502644777297974,
      "eval_loss": 1.2310103178024292,
      "eval_runtime": 9.9985,
      "eval_samples_per_second": 56.708,
      "eval_steps_per_second": 14.202,
      "step": 561
    },
    {
      "epoch": 1.0,
      "step": 561,
      "total_flos": 216400733375928.0,
      "train_loss": 1.3082047596759763,
      "train_runtime": 136.3321,
      "train_samples_per_second": 16.438,
      "train_steps_per_second": 4.115
    }
  ],
  "logging_steps": 500,
  "max_steps": 561,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 216400733375928.0,
  "trial_name": null,
  "trial_params": null
}