metadata
dataset_info:
features:
- name: input_ids
list: int32
- name: labels
list: int32
splits:
- name: train
num_bytes: 3015800064
num_examples: 1503758
download_size: 438490334
dataset_size: 3015800064
configs:
- config_name: default
data_files:
- split: train
path: data/train-*