metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: labels
sequence: int64
- name: input_values
sequence: float32
splits:
- name: train
num_bytes: 12101560609
num_examples: 50670
download_size: 11891879164
dataset_size: 12101560609