{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 783,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.92,
      "learning_rate": 1.8071519795657726e-05,
      "loss": 0.3211,
      "step": 500
    },
    {
      "epoch": 3.0,
      "step": 783,
      "total_flos": 38260492861440.0,
      "train_loss": 0.30118698178579983,
      "train_runtime": 460.6288,
      "train_samples_per_second": 6.799,
      "train_steps_per_second": 1.7
    }
  ],
  "logging_steps": 500,
  "max_steps": 783,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 38260492861440.0,
  "trial_name": null,
  "trial_params": null
}