---
dataset_info:
  features:
  - name: subject_qid
    dtype: string
  - name: subject_text
    dtype: string
  - name: answer_qid
    dtype: string
  - name: answer_text
    dtype: string
  - name: question
    dtype: string
  - name: subject_chunks
    sequence: int32
  - name: subject_num_chunks
    dtype: int32
  - name: answer_chunks
    sequence: int32
  - name: answer_num_chunks
    dtype: int32
  - name: shared_chunks
    sequence: int32
  - name: num_shared_chunks
    dtype: int32
  - name: subject_aliases
    sequence: string
  - name: answer_aliases
    sequence: string
  splits:
  - name: train
    num_bytes: 1837258427
    num_examples: 75207
  download_size: 1375012289
  dataset_size: 1837258427
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---