Land Ccw 37 04 15 62

GPTClaudeDeepSeek··▲ 1,145 copies·updated 2026-07-14

land-ccw-37-04-15-62.prompt

leoscorpius-7b.Q5_0.gguf

---

# Task path
/arc-dataset-collection/dataset/arc-dataset-diva/data/land_ccw/land_ccw_37_04_15_62.json

# Task json
{'train': [{'input': [[3, 7]], 'output': [[7], [3]]}, {'input': [[0, 4]], 'output': [[4], [0]]}], 'test': [{'input': [[1, 5]], 'output': [[5], [1]]}, {'input': [[6, 2]], 'output': [[2], [6]]}]}



# Misc
model=local-model
temperature=0.0
token_limit=4096


# System content:
You solve puzzles by transforming the input into the output. You are expert at the PGM (Portable Graymap) format.


# User content: 741 bytes
# Transformations

## Transformation A - Input

fill the variables

This prompt has 4 variables. Pro fills them into a ready-to-paste prompt for you — no manual find-and-replace.

{'train': [{'input': [[3, 7]], 'output': [[7], [3]]}{'input': [[0, 4]], 'output': [[4], [0]]}{'input': [[1, 5]], 'output': [[5], [1]]}{'input': [[6, 2]], 'output': [[2], [6]]}

Unlock with Pro →

when to use it

Community prompt sourced from the open-source GitHub repo neoneye/arc-prompt (Apache-2.0). A "Land Ccw 37 04 15 62" style prompt — adapt the placeholders and specifics to your task. Imported as-is and not independently retested here, so check the output before relying on it.