metadata
dataset_info:
features:
- name: tokens
sequence:
sequence: int16
- name: mask
sequence:
sequence: bool
splits:
- name: train
num_bytes: 579449352
num_examples: 8313
download_size: 375690299
dataset_size: 579449352
configs:
- config_name: default
data_files:
- split: train
path: data/train-*