metadata
dataset_info:
features:
- name: path
dtype: string
- name: task_binary
dtype: binary
splits:
- name: train
num_bytes: 5227901
num_examples: 2010
download_size: 5203305
dataset_size: 5227901
configs:
- config_name: default
data_files:
- split: train
path: data/train-*