---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: title
    dtype: string
  - name: context
    dtype: string
  - name: question
    dtype: string
  - name: answers
    sequence:
    - name: text
      dtype: string
    - name: answer_start
      dtype: int32
  splits:
  - name: train
    num_bytes: 81103194
    num_examples: 87599
  download_size: 19538606
  dataset_size: 81103194
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---