trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 7.874105865522175,
  "global_step": 344,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.02,
      "learning_rate": 0.0,
      "loss": 2.3574,
      "step": 1
    },
    {
      "epoch": 7.87,
      "step": 344,
      "total_flos": 811093500887040.0,
      "train_loss": 0.6403898139332616,
      "train_runtime": 112463.8178,
      "train_samples_per_second": 0.795,
      "train_steps_per_second": 0.003
    }
  ],
  "max_steps": 344,
  "num_train_epochs": 8,
  "total_flos": 811093500887040.0,
  "trial_name": null,
  "trial_params": null
}