metadata
dataset_info:
features:
- name: query
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 71550854
num_examples: 1000000
download_size: 38072158
dataset_size: 71550854
configs:
- config_name: default
data_files:
- split: train
path: data/train-*