metadata
dataset_info:
features:
- name: input_ids
list: int32
- name: attention_mask
list: int8
splits:
- name: train
num_bytes: 42335186656
num_examples: 1043917
- name: test
num_bytes: 2241234664
num_examples: 54948
download_size: 1854596318
dataset_size: 44576421320
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*