---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: retrieved_docs
    list:
    - name: score
      dtype: float64
    - name: text
      dtype: string
  - name: subj
    dtype: string
  - name: prop
    dtype: string
  - name: s_pop
    dtype: int64
  - name: s_pop_sum
    dtype: int64
  - name: s_pop_avg
    dtype: int64
  - name: possible_answers
    list: string
  splits:
  - name: train
    num_bytes: 18061011
    num_examples: 22075
  download_size: 9803849
  dataset_size: 18061011
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---