metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: cluster
dtype: int64
- name: embedding_512
sequence: float32
splits:
- name: train
num_bytes: 2091790625
num_examples: 990526
download_size: 2669870589
dataset_size: 2091790625
configs:
- config_name: default
data_files:
- split: train
path: data/train-*