{
  "results": {
    "arc_pt": {
      "acc": 0.24444444444444444,
      "acc_stderr": 0.012569442967524465,
      "acc_norm": 0.305982905982906,
      "acc_norm_stderr": 0.01347802974882896
    },
    "hellaswag_pt": {
      "acc": 0.35572651424856433,
      "acc_stderr": 0.004983557286529839,
      "acc_norm": 0.42832376205439376,
      "acc_norm_stderr": 0.005151187541296219
    },
    "truthfulqa_pt": {
      "mc1": 0.23857868020304568,
      "mc1_stderr": 0.015192910034567013,
      "mc2": 0.411729137477953,
      "mc2_stderr": 0.014880046850377518
    }
  },
  "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_17066349/step_21084",
    "batch_size": 1,
    "device": "cuda:0",
    "no_cache": false,
    "limit": null,
    "bootstrap_iters": 100000,
    "description_dict": {}
  }
}