metadata
dataset_info:
features:
- name: tokens
sequence: int64
splits:
- name: train
num_bytes: 383569032
num_examples: 792498
- name: test
num_bytes: 95892500
num_examples: 198125
download_size: 45506002
dataset_size: 479461532
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*