# The Independence Prior of SAEs Fragments Visual Concepts

> Полный текст доступен на английском

**Published:** 6 октября 2026 г.
**Author:** Tommaso Mencattini, Giorgos Nikolaou, Donato Crisostomi, Thomas Fel, Francesco Montagna, Emanuele Rodol\`a, Francesco Locatello
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/ru/articles/the-independence-prior-of-saes-fragments-visual-concepts

Sparse Autoencoders (SAEs) decompose model activations into sparse combinations of interpretable dictionary atoms. Although SAEs are grounded in the Linear Representation Hypothesis (LRH), their objective smuggles in an additional prior: concepts across patches are treated as…

> Материал агрегирован из первоисточника: **arXiv — cs.AI**
> [Читать полностью в arXiv — cs.AI](https://arxiv.org/abs/2610.04112)

**Теги:** #arXiv

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*