metadata
dataset_info:
features:
- name: joke
dtype: string
splits:
- name: train
num_bytes: 5998
num_examples: 20
download_size: 3897
dataset_size: 5998
configs:
- config_name: default
data_files:
- split: train
path: data/train-*