metadata
dataset_info:
features:
- name: text
dtype: string
- name: context
dtype: string
- name: question
dtype: string
- name: answers
dtype: string
splits:
- name: train
num_bytes: 23587227
num_examples: 5364
download_size: 8040549
dataset_size: 23587227
configs:
- config_name: default
data_files:
- split: train
path: data/train-*