metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 239328148932
num_examples: 58372726
- name: test
num_bytes: 239107900
num_examples: 58319
download_size: 117076798666
dataset_size: 239567256832
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*