metadata
dataset_info:
features:
- name: bytepair_ids
sequence: int64
splits:
- name: test
num_bytes: 10295552
num_examples: 4358
- name: train
num_bytes: 87275496
num_examples: 36718
- name: validation
num_bytes: 9149336
num_examples: 3760
download_size: 16150551
dataset_size: 106720384
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
- split: train
path: data/train-*
- split: validation
path: data/validation-*