trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "global_step": 297,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 2.241901397705078,
      "eval_runtime": 80.2863,
      "eval_samples_per_second": 12.455,
      "eval_steps_per_second": 1.557,
      "step": 297
    },
    {
      "epoch": 1.0,
      "step": 297,
      "total_flos": 5001918049701888.0,
      "train_loss": 2.2107459688026094,
      "train_runtime": 5291.3805,
      "train_samples_per_second": 3.592,
      "train_steps_per_second": 0.056
    }
  ],
  "max_steps": 297,
  "num_train_epochs": 1,
  "total_flos": 5001918049701888.0,
  "trial_name": null,
  "trial_params": null
}