metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: source
dtype: string
splits:
- name: train
num_bytes: 3952073
num_examples: 16068
download_size: 2192061
dataset_size: 3952073
configs:
- config_name: default
data_files:
- split: train
path: data/train-*