{
  "best_metric": 0.6806233275617818,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-16/checkpoint-1588",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 1588,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.8316681981086731,
      "learning_rate": 0.00012166547953015562,
      "loss": 0.5043,
      "step": 1588
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6806233275617818,
      "eval_f1": 0.28505990133897113,
      "eval_loss": 0.4987773299217224,
      "eval_mcc": 0.17307041091798078,
      "eval_precision": 0.5614156835530881,
      "eval_recall": 0.1910271546635183,
      "eval_runtime": 23.4301,
      "eval_samples_per_second": 542.293,
      "eval_steps_per_second": 16.987,
      "step": 1588
    }
  ],
  "logging_steps": 500,
  "max_steps": 12704,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 3908894076300.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7919141559939151,
    "learning_rate": 0.00013904626232017785,
    "num_train_epochs": 8,
    "temperature": 8
  }
}