{
  "best_metric": 0.5459882583170255,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-7/checkpoint-49",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 49,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 0.2983497977256775,
      "learning_rate": 0.00041518968722937865,
      "loss": 0.3003,
      "step": 49
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.5459882583170255,
      "eval_f1": 0.22147651006711408,
      "eval_loss": 0.2723988890647888,
      "eval_precision": 0.7764705882352941,
      "eval_recall": 0.12915851272015655,
      "eval_runtime": 28.4829,
      "eval_samples_per_second": 35.881,
      "eval_steps_per_second": 1.123,
      "step": 49
    }
  ],
  "logging_steps": 500,
  "max_steps": 196,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 4,
  "save_steps": 500,
  "total_flos": 235695197280.0,
  "train_batch_size": 63,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.3714301504000702,
    "learning_rate": 0.0005535862496391715,
    "num_train_epochs": 4,
    "per_device_train_batch_size": 63,
    "temperature": 7
  }
}