---
dataset_info:
  features:
  - name: subset
    dtype: string
  - name: context
    dtype: string
  - name: question
    dtype: string
  - name: answers
    struct:
    - name: answer_start
      sequence: int64
    - name: text
      sequence: string
  - name: id
    dtype: string
  splits:
  - name: train
    num_bytes: 1229206945
    num_examples: 516819
  - name: validation
    num_bytes: 145431771
    num_examples: 58221
  - name: test
    num_bytes: 17356377
    num_examples: 9633
  download_size: 665373005
  dataset_size: 1391995093
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---