{
  "id": 12292327,
  "title": "A Testable Theory of Atomic Features",
  "url": "https://urgent.news/2026/10/05/a-testable-theory-of-atomic-features",
  "topic": "ai",
  "section": "AI",
  "published": "2026-10-05T04:40:24.000Z",
  "source": {
    "name": "arXiv cs.AI",
    "slug": "arxiv-cs-ai",
    "url": "https://arxiv.org/abs/2610.05794v1"
  },
  "original_language": "en",
  "account": null,
  "summary": "We develop and test a theory of language model representations in which there exist atomic features. Our main theoretical insight is that in such a model, sparse dictionaries (e.g., SAEs) of increasing size recover an increasing prefix of the most prevalent atoms in the training data. This \"recovery principle\" yields three testable predictions: many features in small SAEs are shared by all larger…",
  "key_points": [],
  "editors_take": null,
  "illustration": null,
  "coverage": {
    "outlets": 1,
    "also_reported_by": []
  },
  "ai_generated": true,
  "disclaimer": "Summaries, key points and the editor’s take are written by software from other outlets’ reporting and may contain errors — always check the linked original."
}