{
  "best_metric": 0.8371559633027523,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-34/checkpoint-527",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 1054,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 20.85869789123535,
      "learning_rate": 0.0002713055538909492,
      "loss": 1.5596,
      "step": 527
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.8371559633027523,
      "eval_loss": 1.5567338466644287,
      "eval_runtime": 2.8111,
      "eval_samples_per_second": 310.197,
      "eval_steps_per_second": 2.49,
      "step": 527
    },
    {
      "epoch": 2.0,
      "grad_norm": 6.722352027893066,
      "learning_rate": 0.00022608796157579106,
      "loss": 0.6915,
      "step": 1054
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.819954128440367,
      "eval_loss": 1.7321298122406006,
      "eval_runtime": 2.823,
      "eval_samples_per_second": 308.894,
      "eval_steps_per_second": 2.48,
      "step": 1054
    }
  ],
  "logging_steps": 500,
  "max_steps": 3689,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 7,
  "save_steps": 500,
  "total_flos": 16141525436100.0,
  "train_batch_size": 128,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.2366804151203985,
    "learning_rate": 0.0003165231462061075,
    "num_train_epochs": 7,
    "temperature": 22
  }
}