metadata
dataset_info:
features:
- name: bytepair
sequence: int64
- name: index
dtype: int64
splits:
- name: train
num_bytes: 1835008
num_examples: 65536
download_size: 589943
dataset_size: 1835008
configs:
- config_name: default
data_files:
- split: train
path: data/train-*