metadata
dataset_info:
features:
- name: human
dtype: string
splits:
- name: train
num_bytes: 15989722
num_examples: 21600
download_size: 9641502
dataset_size: 15989722
configs:
- config_name: default
data_files:
- split: train
path: data/train-*