metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: token_type_ids
sequence: int8
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 794464
num_examples: 2684
- name: test
num_bytes: 184112
num_examples: 622
download_size: 221556
dataset_size: 978576
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*