metadata
dataset_info:
features:
- name: embedding
sequence: float64
- name: score
dtype: float64
splits:
- name: train
num_bytes: 828416160
num_examples: 25272
download_size: 644889358
dataset_size: 828416160
configs:
- config_name: default
data_files:
- split: train
path: data/train-*