---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: title
    dtype: string
  - name: context
    dtype: string
  - name: question
    dtype: string
  - name: answers
    struct:
    - name: answer_start
      sequence: int64
    - name: text
      sequence: string
  - name: masked_query
    dtype: string
  - name: query_embedding
    sequence: float64
  - name: random_answer
    dtype: string
  - name: similar_answer
    dtype: string
  - name: similar_answer_v2
    dtype: string
  - name: answer
    dtype: string
  - name: __index_level_0__
    dtype: int64
  splits:
  - name: train
    num_bytes: 923830110
    num_examples: 97748
  download_size: 591722376
  dataset_size: 923830110
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---