metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 1122623923.6631918
num_examples: 7942534
- name: test
num_bytes: 1413432.9467940482
num_examples: 10000
download_size: 649717010
dataset_size: 1124037356.6099858
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*