---
dataset_info:
  features:
  - 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: 12472788.218033005
    num_examples: 7999
  - name: validation
    num_bytes: 1560852.7323729263
    num_examples: 1001
  - name: test
    num_bytes: 1559293.4389339923
    num_examples: 1000
  download_size: 8019202
  dataset_size: 15592934.389339924
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
  - split: validation
    path: data/validation-*
  - split: test
    path: data/test-*
---