{
  "best_metric": 0.7749510763209393,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-31/checkpoint-96",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 96,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.4516360759735107,
      "learning_rate": 0.00019035656138769916,
      "loss": 0.4942,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7749510763209393,
      "eval_f1": 0.8037542662116042,
      "eval_loss": 0.4383643567562103,
      "eval_precision": 0.7125567322239031,
      "eval_recall": 0.9217221135029354,
      "eval_runtime": 25.1128,
      "eval_samples_per_second": 40.696,
      "eval_steps_per_second": 1.274,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 864,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7140481525355762,
    "learning_rate": 0.00021415113156116159,
    "num_train_epochs": 9,
    "temperature": 6
  }
}