trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.7027027027027026,
  "eval_steps": 500,
  "global_step": 500,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.05376909673213959,
      "eval_runtime": 8922.8969,
      "eval_samples_per_second": 0.162,
      "eval_steps_per_second": 0.02,
      "step": 185
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.04495662450790405,
      "eval_runtime": 7343.0062,
      "eval_samples_per_second": 0.197,
      "eval_steps_per_second": 0.025,
      "step": 370
    },
    {
      "epoch": 2.7,
      "learning_rate": 1.9819819819819822e-06,
      "loss": 0.2417,
      "step": 500
    }
  ],
  "logging_steps": 500,
  "max_steps": 555,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 1079759868002304.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}