{
  "best_metric": 0.7886497064579256,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-21/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": 1.274248719215393,
      "learning_rate": 0.00018849707699065689,
      "loss": 0.5008,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7886497064579256,
      "eval_f1": 0.807142857142857,
      "eval_loss": 0.4340507984161377,
      "eval_precision": 0.7422003284072249,
      "eval_recall": 0.8845401174168297,
      "eval_runtime": 30.5806,
      "eval_samples_per_second": 33.42,
      "eval_steps_per_second": 1.046,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 480,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 5,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7341174287180767,
    "learning_rate": 0.0002356213462383211,
    "num_train_epochs": 5,
    "temperature": 30
  }
}