metadata
dataset_info:
features:
- name: '0'
dtype: string
- name: embeddings
sequence: float32
splits:
- name: train
num_bytes: 13723139
num_examples: 7820
download_size: 17876135
dataset_size: 13723139
configs:
- config_name: default
data_files:
- split: train
path: data/train-*