{
  "best_metric": 0.6728317330395089,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-14/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": 1.0097990036010742,
      "learning_rate": 0.00014225432321746245,
      "loss": 0.6009,
      "step": 1588
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.6728317330395089,
      "eval_f1": 0.4162336750456397,
      "eval_loss": 0.5847160816192627,
      "eval_mcc": 0.20724185361309674,
      "eval_precision": 0.5135135135135135,
      "eval_recall": 0.3499409681227863,
      "eval_runtime": 23.7837,
      "eval_samples_per_second": 534.232,
      "eval_steps_per_second": 16.734,
      "step": 1588
    }
  ],
  "logging_steps": 500,
  "max_steps": 11116,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 3908894076300.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9737159843938835,
    "learning_rate": 0.00016596337708703953,
    "num_train_epochs": 7,
    "temperature": 10
  }
}