trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 528,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4932644367218018,
      "learning_rate": 0.0,
      "loss": 0.226,
      "step": 528
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.963231640663818,
      "eval_f1": 0.8287262872628726,
      "eval_loss": 0.15753257274627686,
      "eval_precision": 0.776536312849162,
      "eval_recall": 0.8884369552585706,
      "eval_runtime": 5.4112,
      "eval_samples_per_second": 172.791,
      "eval_steps_per_second": 2.772,
      "step": 528
    },
    {
      "epoch": 1.0,
      "step": 528,
      "total_flos": 230610745968444.0,
      "train_loss": 0.2260448426911325,
      "train_runtime": 127.3009,
      "train_samples_per_second": 66.276,
      "train_steps_per_second": 4.148
    }
  ],
  "logging_steps": 500,
  "max_steps": 528,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 1,
  "save_steps": 500,
  "total_flos": 230610745968444.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": null
}