{
  "best_metric": 0.5391389432485323,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-6/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": 0.47783181071281433,
      "learning_rate": 0.00014230986748683228,
      "loss": 0.0575,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5391389432485323,
      "eval_f1": 0.19487179487179485,
      "eval_loss": 0.03508628159761429,
      "eval_precision": 0.7702702702702703,
      "eval_recall": 0.11154598825831702,
      "eval_runtime": 29.4088,
      "eval_samples_per_second": 34.751,
      "eval_steps_per_second": 1.088,
      "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": 0.026597019089590446,
    "learning_rate": 0.00016009860092268632,
    "num_train_epochs": 9,
    "temperature": 12
  }
}