metadata
dataset_info:
features:
- name: text
dtype: string
- name: frozen_embeddings
sequence: float64
splits:
- name: train
num_bytes: 1636290725
num_examples: 93695
download_size: 305363682
dataset_size: 1636290725
configs:
- config_name: default
data_files:
- split: train
path: data/train-*