metadata
dataset_info:
config_name: en
features:
- name: user
dtype: string
- name: <image1>
dtype: string
- name: <image2>
dtype: string
- name: assistant
dtype: string
- name: score
dtype: int64
- name: conversation_id
dtype: string
splits:
- name: train
num_bytes: 232617886
num_examples: 124405
download_size: 69079248
dataset_size: 232617886
configs:
- config_name: en
data_files:
- split: train
path: en/train-*