metadata
dataset_info:
features:
- name: text
dtype: string
- name: question
dtype: string
- name: label
dtype: int64
splits:
- name: train
num_bytes: 8523200
num_examples: 20107
download_size: 4759406
dataset_size: 8523200
configs:
- config_name: default
data_files:
- split: train
path: data/train-*