trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.336448598130841,
  "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_accuracy": 0.8134739308728763,
      "eval_f1": 0.3286454478164323,
      "eval_loss": 0.910315215587616,
      "eval_precision": 0.2187192118226601,
      "eval_recall": 0.6607142857142857,
      "eval_runtime": 1.6161,
      "eval_samples_per_second": 28.464,
      "eval_steps_per_second": 28.464,
      "step": 214
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.838605741066198,
      "eval_f1": 0.3566146261298274,
      "eval_loss": 0.8989921808242798,
      "eval_precision": 0.246311010215664,
      "eval_recall": 0.6458333333333334,
      "eval_runtime": 1.5526,
      "eval_samples_per_second": 29.627,
      "eval_steps_per_second": 29.627,
      "step": 428
    },
    {
      "epoch": 2.34,
      "learning_rate": 4.42367601246106e-06,
      "loss": 0.7629,
      "step": 500
    }
  ],
  "logging_steps": 500,
  "max_steps": 642,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 95123329040250.0,
  "train_batch_size": 1,
  "trial_name": null,
  "trial_params": null
}