---
dataset_info:
  features:
  - name: id
    dtype: int64
  - name: description
    dtype: string
  - name: photos
    large_list: image
  splits:
  - name: train
    num_bytes: 403295816
    num_examples: 6239
  download_size: 399709915
  dataset_size: 403295816
configs:
- config_name: default
  data_files:
  - split: train
    path: data/train-*
license: apache-2.0
task_categories:
- image-to-text
- visual-question-answering
language:
- ru
multilinguality:
- monolingual
tags:
- meme
- captions
- sft
- qa
---

# text_meme

Соскрапено с отличного Telegram канала [текстовые мемы](https://t.me/text_meme).

## Dataset Description

- **Homepage** https://t.me/text_meme
- **Homepage** https://vk.com/textmeme