{
  "best_metric": 0.5168316831683168,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-9/checkpoint-396",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 396,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.3266456127166748,
      "learning_rate": 0.00047485708344336834,
      "loss": 0.1574,
      "step": 396
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5168316831683168,
      "eval_f1": 0.18120805369127518,
      "eval_loss": 0.15217570960521698,
      "eval_precision": 0.5869565217391305,
      "eval_recall": 0.10714285714285714,
      "eval_runtime": 14.9861,
      "eval_samples_per_second": 33.698,
      "eval_steps_per_second": 1.068,
      "step": 396
    }
  ],
  "logging_steps": 500,
  "max_steps": 2772,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 777843283140.0,
  "train_batch_size": 24,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.17422929745790272,
    "learning_rate": 0.0005539999306839298,
    "num_train_epochs": 7,
    "per_device_train_batch_size": 24,
    "temperature": 12
  }
}