metadata
dataset_info:
features:
- name: input_ids
sequence: int32
- name: attention_mask
sequence: int8
- name: labels
sequence: int64
- name: length
dtype: int64
splits:
- name: train
num_bytes: 186980559
num_examples: 151994
download_size: 41026135
dataset_size: 186980559