{
  "best_metric": 0.6491421375728003,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-5/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": 0.13218343257904053,
      "learning_rate": 0.00015920412041008642,
      "loss": 0.0241,
      "step": 1588
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6491421375728003,
      "eval_f1": 0.01109139307897072,
      "eval_loss": 0.01951172575354576,
      "eval_mcc": -0.07598752829291255,
      "eval_precision": 0.09157509157509157,
      "eval_recall": 0.0059031877213695395,
      "eval_runtime": 23.3183,
      "eval_samples_per_second": 544.893,
      "eval_steps_per_second": 17.068,
      "step": 1588
    }
  ],
  "logging_steps": 500,
  "max_steps": 14292,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 9,
  "save_steps": 500,
  "total_flos": 3908894076300.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.020625425582708967,
    "learning_rate": 0.00017910463546134722,
    "num_train_epochs": 9,
    "temperature": 19
  }
}