{
  "best_metric": 0.8428062719060604,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-stsb/run-0/checkpoint-720",
  "epoch": 5.0,
  "eval_steps": 500,
  "global_step": 720,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_loss": 0.9918314814567566,
      "eval_pearson": 0.7966375426002391,
      "eval_runtime": 0.8021,
      "eval_samples_per_second": 1870.124,
      "eval_spearmanr": 0.7909051383387676,
      "eval_steps_per_second": 117.194,
      "step": 144
    },
    {
      "epoch": 2.0,
      "eval_loss": 0.8358427882194519,
      "eval_pearson": 0.8281295598730645,
      "eval_runtime": 0.826,
      "eval_samples_per_second": 1815.925,
      "eval_spearmanr": 0.8254389941774403,
      "eval_steps_per_second": 113.798,
      "step": 288
    },
    {
      "epoch": 3.0,
      "eval_loss": 0.7649794220924377,
      "eval_pearson": 0.8375909620622599,
      "eval_runtime": 0.8159,
      "eval_samples_per_second": 1838.53,
      "eval_spearmanr": 0.8346396296700553,
      "eval_steps_per_second": 115.215,
      "step": 432
    },
    {
      "epoch": 3.47,
      "grad_norm": 14.258247375488281,
      "learning_rate": 1.3217957030592118e-05,
      "loss": 1.0074,
      "step": 500
    },
    {
      "epoch": 4.0,
      "eval_loss": 0.6654197573661804,
      "eval_pearson": 0.8417994404614288,
      "eval_runtime": 0.8349,
      "eval_samples_per_second": 1796.689,
      "eval_spearmanr": 0.835660530728647,
      "eval_steps_per_second": 112.592,
      "step": 576
    },
    {
      "epoch": 5.0,
      "eval_loss": 0.6604835391044617,
      "eval_pearson": 0.8428062719060604,
      "eval_runtime": 0.8131,
      "eval_samples_per_second": 1844.83,
      "eval_spearmanr": 0.8363367917240269,
      "eval_steps_per_second": 115.609,
      "step": 720
    }
  ],
  "logging_steps": 500,
  "max_steps": 720,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 22173922656444.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 4.325876846375602e-05,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 4,
    "seed": 34
  }
}