metadata
dataset_info:
features:
- name: chunks
dtype: string
splits:
- name: train
num_bytes: 19406680850
num_examples: 22300752
download_size: 11345406685
dataset_size: 19406680850
configs:
- config_name: default
data_files:
- split: train
path: data/train-*