{
  "best_metric": 0.7661448140900196,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-33/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": 3.7092182636260986,
      "learning_rate": 0.000176452045162556,
      "loss": 0.5741,
      "step": 96
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.7661448140900196,
      "eval_f1": 0.8032921810699588,
      "eval_loss": 0.48392367362976074,
      "eval_precision": 0.6931818181818182,
      "eval_recall": 0.9549902152641878,
      "eval_runtime": 25.4757,
      "eval_samples_per_second": 40.117,
      "eval_steps_per_second": 1.256,
      "step": 96
    }
  ],
  "logging_steps": 500,
  "max_steps": 768,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 8,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.9338649735083602,
    "learning_rate": 0.0002016594801857783,
    "num_train_epochs": 8,
    "temperature": 4
  }
}