metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: percent
dtype: float64
splits:
- name: train
num_bytes: 16400710
num_examples: 7473
download_size: 4069170
dataset_size: 16400710
configs:
- config_name: default
data_files:
- split: train
path: data/train-*