{
  "best_metric": 0.7152641878669276,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-13/checkpoint-192",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 192,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.2103345394134521,
      "learning_rate": 0.0007303245817667495,
      "loss": 0.5045,
      "step": 192
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7152641878669276,
      "eval_f1": 0.7390134529147983,
      "eval_loss": 0.4673319160938263,
      "eval_mcc": 0.43784064067818973,
      "eval_precision": 0.6821192052980133,
      "eval_recall": 0.8062622309197651,
      "eval_runtime": 66.234,
      "eval_samples_per_second": 15.43,
      "eval_steps_per_second": 0.483,
      "step": 192
    }
  ],
  "logging_steps": 500,
  "max_steps": 768,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 16,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7599505594326271,
    "learning_rate": 0.0009737661090223328,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 16,
    "temperature": 27
  }
}