metadata
dataset_info:
features:
- name: sequence
dtype: string
splits:
- name: val
num_bytes: 39829408128
num_examples: 304032
- name: train
num_bytes: 317608455672
num_examples: 2424418
- name: test
num_bytes: 40119319980
num_examples: 306245
download_size: 184540835138
dataset_size: 397557183780
configs:
- config_name: default
data_files:
- split: val
path: data/val-*
- split: train
path: data/train-*
- split: test
path: data/test-*