metadata
dataset_info:
features:
- name: labels
sequence: int64
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
splits:
- name: train
num_bytes: 54115315.2
num_examples: 32688
- name: test
num_bytes: 6012812.8
num_examples: 3632
download_size: 20544882
dataset_size: 60128128