trainer_state.json

{
  "best_metric": null,
  "best_model_checkpoint": null,
  "epoch": 2.9957686882933707,
  "eval_steps": 500,
  "global_step": 1062,
  "is_hyper_param_search": false,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.09817185997962952,
      "eval_mse": 0.09817185997962952,
      "eval_pearsonr": 0.39754295469203044,
      "eval_runtime": 16.9607,
      "eval_samples_per_second": 167.092,
      "eval_spearmanr": 0.40639190714439083,
      "eval_steps_per_second": 20.931,
      "step": 354
    },
    {
      "epoch": 1.41,
      "learning_rate": 4.119371727748691e-06,
      "loss": 0.1166,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.08804947137832642,
      "eval_mse": 0.08804947882890701,
      "eval_pearsonr": 0.48670442010425546,
      "eval_runtime": 16.8904,
      "eval_samples_per_second": 167.788,
      "eval_spearmanr": 0.48893639844454545,
      "eval_steps_per_second": 21.018,
      "step": 709
    },
    {
      "epoch": 2.82,
      "learning_rate": 4.5445026178010473e-07,
      "loss": 0.0903,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.09168503433465958,
      "eval_mse": 0.09168502688407898,
      "eval_pearsonr": 0.48928741659848096,
      "eval_runtime": 16.7916,
      "eval_samples_per_second": 168.774,
      "eval_spearmanr": 0.4923729823700353,
      "eval_steps_per_second": 21.141,
      "step": 1062
    },
    {
      "epoch": 3.0,
      "step": 1062,
      "total_flos": 3498948102716928.0,
      "train_loss": 0.10231328010559082,
      "train_runtime": 760.799,
      "train_samples_per_second": 44.724,
      "train_steps_per_second": 1.396
    }
  ],
  "logging_steps": 500,
  "max_steps": 1062,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3498948102716928.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}