metadata
dataset_info:
features:
- name: paragraph
dtype: string
- name: question
dtype: string
- name: answers
struct:
- name: text
sequence: string
splits:
- name: train
num_bytes: 402614
num_examples: 388
download_size: 267848
dataset_size: 402614
configs:
- config_name: default
data_files:
- split: train
path: data/train-*