metadata
dataset_info:
features:
- name: text
dtype: string
- name: questions
sequence: string
- name: answers
sequence:
sequence: string
- name: document_id
dtype: string
- name: split
dtype: string
splits:
- name: train
num_bytes: 508515599
num_examples: 1572
download_size: 288012667
dataset_size: 508515599
configs:
- config_name: default
data_files:
- split: train
path: data/train-*