Metraipolis › archive › dc_9528a751d994
Increasing Skill Level Recruits Deeper Attention Layers in a Frozen Chess Transformer — arXiv:2609.23917
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.23917
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T07:08:37.819Z
- SHA-256 of the archived text
a7094d4aaba58a67022f0ee9d6dd82d81b98ff6d9be45b96a99999602be07d06- Stable URL of this page
https://metraipolis.com/documento/dc_9528a751d994
How to cite
Increasing Skill Level Recruits Deeper Attention Layers in a Frozen Chess Transformer — arXiv:2609.23917 — arXiv (en), https://arxiv.org/abs/2609.23917 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_9528a751d994 (Metraipolis archive dc_9528a751d994, sha256 a7094d4aaba58a67022f0ee9d6dd82d81b98ff6d9be45b96a99999602be07d06)
BibTeX
@misc{dc9528a751d994,
title = {Increasing Skill Level Recruits Deeper Attention Layers in a Frozen Chess Transformer — arXiv:2609.23917},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.23917},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_9528a751d994; sha256 a7094d4aaba58a67022f0ee9d6dd82d81b98ff6d9be45b96a99999602be07d06},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.23917. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Chess involves complex reasoning in a deterministic environment, which makes it a useful setting for studying the mechanisms of computation inside transformers. The Maia-3 chess transformer takes Elo, a measure of competitive chess skill, as an input to the pre-trained network, so we can vary the skill the network is conditioned on with no change to its weights. Here we investigate how turning this skill dial affects self-attention. Ablating every attention head at every Elo from 700 to 2500, we find 1) increasing skill pushes the causal center of mass of the computation deeper, monotonically, for every chess piece and move type we measured; 2) the depth migration is much greater for specific tactics, especially knight forks, than for other move types; 3) the migration consists of deeper heads getting recruited for more specialized computations while one shared shallow head keeps a roughly constant contribution. These results may shed light on how conditioning inputs redistribute computation in larger transformers. Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-22 in cs.LG). Categorie arXiv: Artificial Intelligence (cs.AI) ; Machine Learning (cs.LG). Pagina del lavoro: https://arxiv.org/abs/2609.23917 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-09T06:58:09.606Z dalla pagina https://arxiv.org/catchup/cs.LG/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).