metadata
dataset_info:
features:
- name: bytepair_ids
sequence: int64
splits:
- name: test
num_bytes: 5156396
num_examples: 2891
- name: train
num_bytes: 43707492
num_examples: 23767
- name: validation
num_bytes: 4581844
num_examples: 2461
download_size: 9438046
dataset_size: 53445732
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*
- split: validation
path: data/validation-*