metadata
dataset_info:
features:
- name: embedding
sequence: float32
splits:
- name: test
num_bytes: 30760000
num_examples: 10000
download_size: 36966542
dataset_size: 30760000
configs:
- config_name: default
data_files:
- split: test
path: data/test-*
