{
  "best_metric": 0.5156555772994129,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-17/checkpoint-55",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 55,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 1.0903576612472534,
      "learning_rate": 4.005913685397949e-05,
      "loss": 0.6291,
      "step": 55
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5156555772994129,
      "eval_f1": 0.10488245931283906,
      "eval_loss": 0.618636429309845,
      "eval_precision": 0.6904761904761905,
      "eval_recall": 0.05675146771037182,
      "eval_runtime": 28.4346,
      "eval_samples_per_second": 35.942,
      "eval_steps_per_second": 1.125,
      "step": 55
    }
  ],
  "logging_steps": 500,
  "max_steps": 220,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 56,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.8797585777570165,
    "learning_rate": 5.3412182471972654e-05,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 56,
    "temperature": 28
  }
}