---
license: mit
task_categories:
- image-to-text
- text-to-image
language:
- en
pretty_name: simons ARC (abstraction & reasoning corpus) cellular automaton version 3
size_categories:
- 10K<n<100K
configs:
  - config_name: default
    data_files:
      - split: train
        path: data.jsonl
---

# Version 1

The images are run length encoded (RLE).

The image sizes are between 4 and 10 pixels.

Cellular automaton types:
- gameoflife
- highlife
- serviettes
- cave
- maze

Number of CA steps, range 1-2.

The LLM was not happy about this dataset.

# Version 2

The image sizes are between 4 and 11 pixels.

Number of CA steps = 1.

# Version 3

Disabled nowrap. It's only CAs that wraps around.

The image sizes are between 4 and 11 pixels.

Number of CA steps = 1.