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: 493127488
num_examples: 119924
- name: validation
num_bytes: 27274896
num_examples: 6633
- name: test
num_bytes: 27377696
num_examples: 6658
download_size: 153946164
dataset_size: 547780080