{
  "best_metric": 0.8325688073394495,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-25/checkpoint-527",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1054,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 3.7773447036743164,
      "learning_rate": 0.0004316595412152798,
      "loss": 0.9448,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8325688073394495,
      "eval_loss": 1.1720139980316162,
      "eval_runtime": 2.8085,
      "eval_samples_per_second": 310.481,
      "eval_steps_per_second": 2.492,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 9.26517105102539,
      "learning_rate": 0.00036999389247023985,
      "loss": 0.4379,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.8061926605504587,
      "eval_loss": 1.3765146732330322,
      "eval_runtime": 2.8111,
      "eval_samples_per_second": 310.197,
      "eval_steps_per_second": 2.49,
      "step": 1054
    }
  ],
  "logging_steps": 500,
  "max_steps": 4216,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 16141525436100.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.595362243351062,
    "learning_rate": 0.0004933251899603198,
    "num_train_epochs": 8,
    "temperature": 26
  }
}