{
  "best_metric": 0.8912369135323769,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-27/checkpoint-645",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 645,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 2.374743938446045,
      "learning_rate": 8.115439310915276e-05,
      "loss": 0.333,
      "step": 645
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8912369135323769,
      "eval_f1": 0.16888888888888887,
      "eval_loss": 0.30435287952423096,
      "eval_mcc": 0.19762599967287012,
      "eval_precision": 0.5377358490566038,
      "eval_recall": 0.10017574692442882,
      "eval_runtime": 9.9164,
      "eval_samples_per_second": 520.146,
      "eval_steps_per_second": 16.337,
      "step": 645
    }
  ],
  "logging_steps": 500,
  "max_steps": 5160,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 1586942532600.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9885260340274179,
    "learning_rate": 9.274787783903173e-05,
    "num_train_epochs": 8,
    "temperature": 24
  }
}