{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 39,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 2.0,
      "grad_norm": 10.93592357635498,
      "learning_rate": 1.6285714285714283e-05,
      "loss": 29.2474,
      "step": 26
    }
  ],
  "logging_steps": 26,
  "max_steps": 39,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3738610012520448.0,
  "train_batch_size": 2,
  "trial_name": null,
  "trial_params": null
}