---
language:
- en
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: 24489106
    num_examples: 27045
  - name: valid
    num_bytes: 3337865
    num_examples: 3688
  - name: test
    num_bytes: 10472984
    num_examples: 10570
  download_size: 19442674
  dataset_size: 38299955
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: valid
    path: data/valid-*
  - split: test
    path: data/test-*
---