{
  "best_metric": 0.4729241877256318,
  "best_model_checkpoint": "microsoft/deberta-v3-smallsed_rte-finetuned-rte/run-2/checkpoint-8",
  "epoch": 2.0,
  "eval_steps": 500,
  "global_step": 16,
  "is_hyper_param_search": true,
  "is_local_process_zero": true,
  "is_world_process_zero": true,
  "log_history": [
    {
      "epoch": 1.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.7006650567054749,
      "eval_runtime": 1.7391,
      "eval_samples_per_second": 159.277,
      "eval_steps_per_second": 10.35,
      "step": 8
    },
    {
      "epoch": 2.0,
      "eval_accuracy": 0.4729241877256318,
      "eval_loss": 0.6988851428031921,
      "eval_runtime": 1.7727,
      "eval_samples_per_second": 156.259,
      "eval_steps_per_second": 10.154,
      "step": 16
    }
  ],
  "logging_steps": 500,
  "max_steps": 16,
  "num_input_tokens_seen": 0,
  "num_train_epochs": 2,
  "save_steps": 500,
  "total_flos": 0,
  "train_batch_size": 32,
  "trial_name": null,
  "trial_params": {
    "learning_rate": 1.71484161515299e-05,
    "num_train_epochs": 2,
    "per_device_train_batch_size": 32,
    "seed": 19
  }
}