{
  "results": {
    "drop": {
      "em": 0.03313758389261745,
      "em_stderr": 0.0018330841858875643,
      "f1": 0.08634542785234882,
      "f1_stderr": 0.0022136353860709133
    },
    "coqa": {
      "f1": 0.4557083534540516,
      "f1_stderr": 0.01876948425119881,
      "em": 0.3298333333333334,
      "em_stderr": 0.019473215823053027
    },
    "race": {
      "acc": 0.3263157894736842,
      "acc_stderr": 0.014510987877134932
    }
  },
  "versions": {
    "drop": 1,
    "coqa": 1,
    "race": 1
  },
  "config": {
    "model": "hf-causal-experimental",
    "model_args": "pretrained=bigscience/bloom-1b1,use_accelerate=True",
    "num_fewshot": 0,
    "batch_size": "auto",
    "device": "cuda:0",
    "no_cache": true,
    "limit": null,
    "bootstrap_iters": 100000,
    "description_dict": {}
  }
}