metadata
dataset_info:
features:
- name: sentence
dtype: string
- name: cluster
dtype: int64
- name: embedding_384
sequence: float32
splits:
- name: train
num_bytes: 1584641313
num_examples: 990526
download_size: 2163024841
dataset_size: 1584641313
configs:
- config_name: default
data_files:
- split: train
path: data/train-*