---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: title
    dtype: string
  - name: context
    dtype: string
  - name: question
    dtype: string
  - name: answers
    struct:
    - name: answer_start
      sequence: int64
    - name: text
      sequence: string
  - name: atom
    dtype: string
  splits:
  - name: train
    num_bytes: 5378931
    num_examples: 5000
  download_size: 1016192
  dataset_size: 5378931
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---