metadata
dataset_info:
features:
- name: sequence
dtype: string
splits:
- name: train
num_bytes: 17532254710
num_examples: 4631863
download_size: 6129418210
dataset_size: 17532254710
configs:
- config_name: default
data_files:
- split: train
path: data/train-*