metadata
dataset_info:
features:
- name: input_ids
sequence:
sequence: int32
- name: token_type_ids
sequence:
sequence: int32
- name: attention_mask
sequence:
sequence:
sequence: bool
- name: labels
sequence:
sequence: int32
splits:
- name: train
num_bytes: 5298664000
num_examples: 2450
download_size: 276339066
dataset_size: 5298664000