metadata
dataset_info:
features:
- name: source
dtype: string
- name: question
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 165505362
num_examples: 114891
download_size: 100546546
dataset_size: 165505362
configs:
- config_name: default
data_files:
- split: train
path: data/train-*