{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 582,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.58,
      "learning_rate": 7.0446735395189e-06,
      "loss": 0.3528,
      "step": 500
    },
    {
      "epoch": 3.0,
      "step": 582,
      "total_flos": 3.6081563868758016e+17,
      "train_loss": 0.3140656472891057,
      "train_runtime": 90.7552,
      "train_samples_per_second": 51.303,
      "train_steps_per_second": 6.413
    }
  ],
  "logging_steps": 500,
  "max_steps": 582,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3.6081563868758016e+17,
  "trial_name": null,
  "trial_params": null
}