metadata
dataset_info:
features:
- name: tokens
list: int64
splits:
- name: train
num_bytes: 1639200000
num_examples: 200000
download_size: 345833980
dataset_size: 1639200000
configs:
- config_name: default
data_files:
- split: train
path: data/train-*