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": 0.0,
      "loss": 0.3188,
      "step": 1
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.7874919176101685,
      "eval_runtime": 123.3726,
      "eval_samples_per_second": 0.819,
      "eval_steps_per_second": 0.413,
      "step": 1
    },
    {
      "epoch": 1.0,
      "step": 1,
      "total_flos": 5993686630400.0,
      "train_loss": 0.31876903772354126,
      "train_runtime": 358.693,
      "train_samples_per_second": 0.282,
      "train_steps_per_second": 0.003
    }
  ],
  "logging_steps": 5,
  "max_steps": 1,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 5993686630400.0,
  "trial_name": null,
  "trial_params": null
}