---
dataset_info:
  features:
  - name: id
    dtype: string
  - name: prompt
    dtype: string
  - name: text_output
    dtype: string
  - name: image_input
    sequence: image
  - name: image_output
    sequence: image
  - name: meta_task_name
    dtype: string
  - name: sub_task_name
    dtype: string
  splits:
  - name: train
    num_bytes: 5127493276.44
    num_examples: 2160
  download_size: 2270235276
  dataset_size: 5127493276.44
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
---