metadata
dataset_info:
features:
- name: question
dtype: string
- name: citation
sequence: string
- name: intro
struct:
- name: intro_1
sequence: string
- name: intro_2
sequence: string
- name: answer
dtype: string
- name: date
dtype: string
- name: refs
sequence: string
- name: old_docs
sequence: string
splits:
- name: train
num_bytes: 112489116
num_examples: 11281
download_size: 13297123
dataset_size: 112489116
configs:
- config_name: default
data_files:
- split: train
path: data/train-*