{
  "best_metric": 0.5948478579521179,
  "best_model_checkpoint": "BERT-WMM/run-12/checkpoint-268",
  "epoch": 3.0,
  "eval_steps": 500,
  "global_step": 402,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7370892018779343,
      "eval_loss": 0.6198122501373291,
      "eval_runtime": 2.0047,
      "eval_samples_per_second": 1062.477,
      "eval_steps_per_second": 66.841,
      "step": 134
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.7633802816901408,
      "eval_loss": 0.5948478579521179,
      "eval_runtime": 2.0123,
      "eval_samples_per_second": 1058.472,
      "eval_steps_per_second": 66.589,
      "step": 268
    },
    {
      "epoch": 3.0,
      "eval_accuracy": 0.7610328638497652,
      "eval_loss": 0.6659887433052063,
      "eval_runtime": 2.1209,
      "eval_samples_per_second": 1004.273,
      "eval_steps_per_second": 63.18,
      "step": 402
    }
  ],
  "logging_steps": 500,
  "max_steps": 402,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 64,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 2.74081157159569e-05,
    "num_train_epochs": 3,
    "per_device_train_batch_size": 64,
    "weight_decay": 0.08087690849456843
  }
}