metadata
dataset_info:
features:
- name: question
dtype: string
- name: field
dtype: string
- name: relevant
dtype: string
splits:
- name: train
num_bytes: 28941457
num_examples: 63271
download_size: 11885508
dataset_size: 28941457
configs:
- config_name: default
data_files:
- split: train
path: data/train-*