metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
splits:
- name: train
num_bytes: 7322130.278739297
num_examples: 4940
- name: test
num_bytes: 813734.7212607032
num_examples: 549
download_size: 697740
dataset_size: 8135865
configs:
- config_name: default
data_files:
- split: train
path: data/train-*
- split: test
path: data/test-*