metadata
dataset_info:
features:
- name: chunks
dtype: string
splits:
- name: train
num_bytes: 19538053723
num_examples: 10204012
download_size: 11372136621
dataset_size: 19538053723
configs:
- config_name: default
data_files:
- split: train
path: data/train-*