{
  "best_metric": 0.70298769771529,
  "best_model_checkpoint": "tiny-bert-sst2-distilled/run-25/checkpoint-214",
  "epoch": 1.0,
  "eval_steps": 500,
  "global_step": 214,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "grad_norm": 6.952327728271484,
      "learning_rate": 0.00011066722418814963,
      "loss": 0.4787,
      "step": 214
    },
    {
      "epoch": 1.0,
      "eval_accuracy": 0.70298769771529,
      "eval_f1": 0.409778812572759,
      "eval_loss": 0.45047906041145325,
      "eval_mcc": 0.26252945389148746,
      "eval_precision": 0.6068965517241379,
      "eval_recall": 0.3093145869947276,
      "eval_runtime": 3.1633,
      "eval_samples_per_second": 539.63,
      "eval_steps_per_second": 17.071,
      "step": 214
    }
  ],
  "logging_steps": 500,
  "max_steps": 3424,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 16,
  "save_steps": 500,
  "total_flos": 524775664440.0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "alpha": 0.7610591791686496,
    "learning_rate": 0.00011804503913402627,
    "num_train_epochs": 16,
    "temperature": 20
  }
}