{
  "results": {
    "arc_pt": {
      "acc": 0.2760683760683761,
      "acc_stderr": 0.0130752454127398,
      "acc_norm": 0.30854700854700856,
      "acc_norm_stderr": 0.01350935904916782
    },
    "hellaswag_pt": {
      "acc": 0.37793910499512406,
      "acc_stderr": 0.005047467981098971,
      "acc_norm": 0.47513273377397336,
      "acc_norm_stderr": 0.005198504535035831
    },
    "truthfulqa_pt": {
      "mc1": 0.24746192893401014,
      "mc1_stderr": 0.015382646812261825,
      "mc2": 0.39066264426629793,
      "mc2_stderr": 0.014567795673718856
    }
  },
  "versions": {
    "arc_pt": 0,
    "hellaswag_pt": 1,
    "truthfulqa_pt": 1
  },
  "config": {
    "model": "hf-auto",
    "model_args": "pretrained=/lustre/mlnvme/data/asen_hpc-mula/checkpoints-llama/slurm_job_17782345/step_42164",
    "batch_size": 1,
    "device": "cuda:0",
    "no_cache": false,
    "limit": null,
    "bootstrap_iters": 100000,
    "description_dict": {}
  }
}