metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: labels
sequence: int64
- name: input_values
sequence: float32
splits:
- name: train
num_bytes: 13067825026.494074
num_examples: 50545
download_size: 11882046988
dataset_size: 13067825026.494074