trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "learning_rate": 0,
      "loss": 0.1523,
      "step": 1
    },
    {
      "epoch": 1.0,
      "step": 1,
      "total_flos": 732627468288.0,
      "train_loss": 0.15234375,
      "train_runtime": 123.9736,
      "train_samples_per_second": 0.008,
      "train_steps_per_second": 0.008
    }
  ],
  "logging_steps": 1.0,
  "max_steps": 1,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 732627468288.0,
  "trial_name": null,
  "trial_params": null
}