metadata
dataset_info:
features:
- name: embeddings
sequence: float32
splits:
- name: train
num_bytes: 340359428
num_examples: 55361
download_size: 373070400
dataset_size: 340359428
configs:
- config_name: default
data_files:
- split: train
path: data/train-*