metadata
dataset_info:
features:
- name: question
dtype: string
- name: answers
struct:
- name: text
sequence: string
splits:
- name: train
num_bytes: 567287
num_examples: 4712
download_size: 288285
dataset_size: 567287
configs:
- config_name: default
data_files:
- split: train
path: data/train-*