{
  "best_metric": 0.512720156555773,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-2/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": 0.3969263434410095,
      "learning_rate": 3.130245712253287e-05,
      "loss": 0.081,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.512720156555773,
      "eval_f1": 0.08118081180811808,
      "eval_loss": 0.03750595450401306,
      "eval_precision": 0.7096774193548387,
      "eval_recall": 0.043052837573385516,
      "eval_runtime": 25.4915,
      "eval_samples_per_second": 40.092,
      "eval_steps_per_second": 1.255,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 288,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 3,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.022604047006827055,
    "learning_rate": 4.6953685683799313e-05,
    "num_train_epochs": 3,
    "temperature": 5
  }
}