---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: title
    dtype: string
  - name: context
    dtype: string
  - name: question
    dtype: string
  - name: answer_text
    dtype: string
  - name: answer_start
    dtype: float64
  - name: language
    dtype: string
  - name: negative_samples
    sequence: string
  - name: __index_level_0__
    dtype: int64
  - name: search_result
    sequence: string
  - name: answer
    dtype: string
  - name: extracted_ref_numbers
    sequence: int64
  splits:
  - name: train
    num_bytes: 5343841
    num_examples: 497
  download_size: 553073
  dataset_size: 5343841
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---