{
  "best_metric": 0.541095890410959,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-4/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": 0.20222608745098114,
      "learning_rate": 0.00010108430383313451,
      "loss": 0.1454,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.541095890410959,
      "eval_f1": 0.20642978003384094,
      "eval_loss": 0.11932022124528885,
      "eval_precision": 0.7625,
      "eval_recall": 0.11937377690802348,
      "eval_runtime": 30.5181,
      "eval_samples_per_second": 33.488,
      "eval_steps_per_second": 1.049,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 672,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.14091216322626776,
    "learning_rate": 0.00011793168780532361,
    "num_train_epochs": 7,
    "temperature": 26
  }
}