metadata
dataset_info:
features:
- name: file_name
dtype: string
- name: document
dtype: string
- name: prompt
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 941904900
num_examples: 41860
- name: test
num_bytes: 241339155
num_examples: 10580
download_size: 28911306
dataset_size: 1183244055
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*