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": 0.09733370691537857,
      "eval_mse": 0.09733370691537857,
      "eval_pearsonr": 0.4643483401235611,
      "eval_runtime": 11.064,
      "eval_samples_per_second": 256.146,
      "eval_spearmanr": 0.46593444267736933,
      "eval_steps_per_second": 32.086,
      "step": 354
    },
    {
      "epoch": 1.41,
      "learning_rate": 5.895287958115184e-06,
      "loss": 0.0949,
      "step": 500
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.09640779346227646,
      "eval_mse": 0.09640779346227646,
      "eval_pearsonr": 0.46914276472553906,
      "eval_runtime": 11.1464,
      "eval_samples_per_second": 254.252,
      "eval_spearmanr": 0.469247931031464,
      "eval_steps_per_second": 31.849,
      "step": 709
    },
    {
      "epoch": 2.82,
      "learning_rate": 6.596858638743456e-07,
      "loss": 0.0785,
      "step": 1000
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.09511137008666992,
      "eval_mse": 0.09511137008666992,
      "eval_pearsonr": 0.4801628579333256,
      "eval_runtime": 11.2595,
      "eval_samples_per_second": 251.699,
      "eval_spearmanr": 0.47989805106952277,
      "eval_steps_per_second": 31.529,
      "step": 1062
    },
    {
      "epoch": 3.0,
      "step": 1062,
      "total_flos": 3160948901412864.0,
      "train_loss": 0.085811287220813,
      "train_runtime": 622.8809,
      "train_samples_per_second": 54.627,
      "train_steps_per_second": 1.705
    }
  ],
  "logging_steps": 500,
  "max_steps": 1062,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 3160948901412864.0,
  "train_batch_size": 8,
  "trial_name": null,
  "trial_params": null
}