metadata
dataset_info:
features:
- name: image
dtype: image
- name: description
dtype: string
- name: conversations
dtype: string
splits:
- name: train
num_bytes: 135069
num_examples: 1
download_size: 143750
dataset_size: 135069
configs:
- config_name: default
data_files:
- split: train
path: data/train-*