metadata
dataset_info:
features:
- name: answer
dtype: string
- name: cluster
dtype: int32
- name: image
dtype: image
- name: prompt
dtype: string
splits:
- name: li_train
num_bytes: 900073595.5
num_examples: 5820
- name: li_test
num_bytes: 354791631.125
num_examples: 2495
download_size: 1239073417
dataset_size: 1254865226.625
configs:
- config_name: default
data_files:
- split: li_train
path: data/li_train-*
- split: li_test
path: data/li_test-*