---
dataset_info:
  features:
  - name: question
    dtype: string
  - name: org_context
    sequence: string
  - name: org_answer
    sequence: string
  - name: sub_context
    sequence: string
  splits:
  - name: train
    num_bytes: 23823996
    num_examples: 3621
  download_size: 12105798
  dataset_size: 23823996
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---