trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 3,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 1.3603196144104004,
      "eval_runtime": 15.8572,
      "eval_samples_per_second": 0.126,
      "eval_steps_per_second": 0.126,
      "step": 1
    },
    {
      "epoch": 2.0,
      "eval_loss": 1.3479301929473877,
      "eval_runtime": 16.7819,
      "eval_samples_per_second": 0.119,
      "eval_steps_per_second": 0.119,
      "step": 2
    },
    {
      "epoch": 3.0,
      "eval_loss": 1.3436630964279175,
      "eval_runtime": 16.3913,
      "eval_samples_per_second": 0.122,
      "eval_steps_per_second": 0.122,
      "step": 3
    },
    {
      "epoch": 3.0,
      "step": 3,
      "total_flos": 528274836946944.0,
      "train_loss": 0.02127033720413844,
      "train_runtime": 216.5079,
      "train_samples_per_second": 0.028,
      "train_steps_per_second": 0.014
    }
  ],
  "logging_steps": 5,
  "max_steps": 3,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 528274836946944.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}