metadata
license: mit
dataset_info:
features:
- name: tokens
sequence: int64
splits:
- name: validation
num_bytes: 1819903404
num_examples: 3760131
- name: train
num_bytes: 9589225800
num_examples: 19812450
download_size: 1922548946
dataset_size: 11409129204
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*