metadata
dataset_info:
features:
- name: question
dtype: string
- name: answer
dtype: string
- name: source-target
dtype: string
splits:
- name: train
num_bytes: 39547120
num_examples: 117404
- name: test
num_bytes: 2964236
num_examples: 8016
download_size: 31918840
dataset_size: 42511356
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*