---
dataset_info:
  features:
  - name: answers
    sequence: string
  - name: passages
    sequence:
    - name: is_selected
      dtype: int32
    - name: passage_text
      dtype: string
    - name: url
      dtype: string
  - name: query
    dtype: string
  - name: query_id
    dtype: int32
  - name: query_type
    dtype: string
  - name: wellFormedAnswers
    sequence: string
  - name: negatives
    sequence: string
  - name: positive
    dtype: string
  splits:
  - name: train
    num_bytes: 518466869
    num_examples: 100000
  download_size: 260609777
  dataset_size: 518466869
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---