{
  "best_metric": 0.4183935375368045,
  "best_model_checkpoint": "distilbert-base-uncased-finetuned-cola/run-2/checkpoint-2138",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 2138,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 0.23,
      "grad_norm": 13.065943717956543,
      "learning_rate": 9.837377839260441e-06,
      "loss": 0.5777,
      "step": 500
    },
    {
      "epoch": 0.47,
      "grad_norm": 14.142450332641602,
      "learning_rate": 9.226512349986941e-06,
      "loss": 0.5355,
      "step": 1000
    },
    {
      "epoch": 0.7,
      "grad_norm": 2.057891845703125,
      "learning_rate": 8.615646860713443e-06,
      "loss": 0.5373,
      "step": 1500
    },
    {
      "epoch": 0.94,
      "grad_norm": 7.0049896240234375,
      "learning_rate": 8.004781371439942e-06,
      "loss": 0.5423,
      "step": 2000
    },
    {
      "epoch": 1.0,
      "eval_loss": 0.6402870416641235,
      "eval_matthews_correlation": 0.4183935375368045,
      "eval_runtime": 37.1869,
      "eval_samples_per_second": 28.048,
      "eval_steps_per_second": 1.775,
      "step": 2138
    }
  ],
  "logging_steps": 500,
  "max_steps": 8552,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 32605608297312.0,
  "train_batch_size": 4,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.0448243328533941e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 4,
    "seed": 17
  }
}