metadata
dataset_info:
features:
- name: input_ids
sequence: int32
splits:
- name: train
num_bytes: 928876604.5181321
num_examples: 7635931
- name: test
num_bytes: 1216454.9476915547
num_examples: 10000
download_size: 524422872
dataset_size: 930093059.4658237
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*