metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
dtype: int64
- name: token_type_ids
sequence: int8
splits:
- name: train
num_bytes: 1139584
num_examples: 9824
download_size: 139270
dataset_size: 1139584