metadata
dataset_info:
features:
- name: feature
dtype: new_feature
splits:
- name: train
num_bytes: 20
num_examples: 5
download_size: 798
dataset_size: 20
configs:
- config_name: default
data_files:
- split: train
path: data/train-*