Ya interesting thought - would be fascinating if generating games w/solutions is part of the training data pipeline. There's been previous work done on on testi

Ya interesting thought - would be fascinating if generating games w/solutions is part of the training data pipeline. There's been previous work done on on testing LLMs on logic puzzles[1][2][3] so they could possibly be building off those ideas to improve performance. [1] https://huggingface.co/papers/2504.00043 https://huggingface.co/papers/2504.00043 [2] https://huggingface.co/blog/yuchenlin/zebra-logic https://huggingface.co/blog/yuchenlin/zebra-logic [3] https://arxiv.org/pdf/2403.12094 https://arxiv.org/pdf/2403.12094