metadata
dataset_info:
features:
- name: a
dtype: int64
- name: b
dtype: int64
splits:
- name: train
num_bytes: 64
num_examples: 4
download_size: 1311
dataset_size: 64
configs:
- config_name: default
data_files:
- split: train
path: data/train-*