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: 3933942
num_examples: 17604
- name: val
num_bytes: 439860
num_examples: 1957
- name: test
num_bytes: 163660
num_examples: 749
download_size: 649747
dataset_size: 4537462