metadata
dataset_info:
features:
- name: tokens
dtype: int64
splits:
- name: train
num_bytes: 7460994112
num_examples: 932624264
- name: validation
num_bytes: 828999352
num_examples: 103624919
download_size: 3290809410
dataset_size: 8289993464
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: validation
path: data/validation-*