metadata
language:
- tr
dataset_info:
features:
- name: image
dtype: image
- name: instruction
dtype: string
- name: gpt4_prediction
dtype: string
- name: instruction_conditioned_caption
dtype: string
splits:
- name: train
num_bytes: 336508697
num_examples: 574
download_size: 335351860
dataset_size: 336508697
configs:
- config_name: default
data_files:
- split: train
path: data/train-*