metadata
dataset_info:
features:
- name: text
dtype: string
- name: embed
sequence: float64
splits:
- name: train
num_bytes: 34325117835
num_examples: 516317
download_size: 18652745837
dataset_size: 34325117835
configs:
- config_name: default
data_files:
- split: train
path: data/train-*