{
  "best_metric": 0.550880626223092,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-18/checkpoint-56",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 56,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.5901849865913391,
      "learning_rate": 0.0003575113771511288,
      "loss": 0.2602,
      "step": 56
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.550880626223092,
      "eval_f1": 0.24630541871921183,
      "eval_loss": 0.2331201732158661,
      "eval_precision": 0.7653061224489796,
      "eval_recall": 0.14677103718199608,
      "eval_runtime": 28.3545,
      "eval_samples_per_second": 36.044,
      "eval_steps_per_second": 1.129,
      "step": 56
    }
  ],
  "logging_steps": 500,
  "max_steps": 280,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 55,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.31316054034254387,
    "learning_rate": 0.000446889221438911,
    "num_train_epochs": 5,
    "per_device_train_batch_size": 55,
    "temperature": 13
  }
}