metadata
dataset_info:
features:
- name: id
dtype: string
- name: document
dtype: string
- name: metadata
dtype: string
- name: embeddings
list: float32
splits:
- name: train
num_bytes: 24637740
num_examples: 5964
download_size: 15895762
dataset_size: 24637740
configs:
- config_name: default
data_files:
- split: train
path: data/train-*