metadata
dataset_info:
features:
- name: tokens
sequence: int64
splits:
- name: train
num_bytes: 45208540
num_examples: 11005
download_size: 7958523
dataset_size: 45208540
configs:
- config_name: default
data_files:
- split: train
path: data/train-*