---
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
  - name: masked_query
    dtype: string
  - name: query_embedding
    sequence: float32
  splits:
  - name: train
    num_bytes: 290349952
    num_examples: 73863
  download_size: 287156848
  dataset_size: 290349952
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---