metadata
dataset_info:
features:
- name: prompt
dtype: string
- name: answer
dtype: string
splits:
- name: train
num_bytes: 50956009
num_examples: 71689
- name: test
num_bytes: 5636410
num_examples: 7832
download_size: 26871987
dataset_size: 56592419
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*