Metraipolis › archive › dc_a7c136cb81eb

Replicating the Geometry of Emotion Representations in a Base Open-Weights Model — arXiv:2609.22208

Source: arXiv (en)
Original: https://arxiv.org/abs/2609.22208
License (as archived): CC0-1.0
Archived: 2026-10-09T06:49:15.798Z
SHA-256 of the archived text: b66b3a12f56c515601c81ac82996a94cf734d4fb66759c1682cb9f1a16c92e36
Stable URL of this page: https://metraipolis.com/documento/dc_a7c136cb81eb

How to cite

Replicating the Geometry of Emotion Representations in a Base Open-Weights Model — arXiv:2609.22208 — arXiv (en), https://arxiv.org/abs/2609.22208 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a7c136cb81eb (Metraipolis archive dc_a7c136cb81eb, sha256 b66b3a12f56c515601c81ac82996a94cf734d4fb66759c1682cb9f1a16c92e36)

BibTeX

@misc{dca7c136cb81eb,
  title = {Replicating the Geometry of Emotion Representations in a Base Open-Weights Model — arXiv:2609.22208},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2609.22208},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a7c136cb81eb; sha256 b66b3a12f56c515601c81ac82996a94cf734d4fb66759c1682cb9f1a16c92e36},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.22208. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Sofroniew et al. (2026) report that emotion concepts in Claude Sonnet 4.5 are represented as vectors whose geometry mirrors human affect psychology. We replicate the representational core of that study on the base pretrained model google/gemma-2-27b, inheriting every disclosed parameter, resolving unspecified steps by disclosed rules, and changing only the subject model. From 205,200 newly generated Claude Sonnet 4.5 stories matching the original corpus design, we extract 171 emotion vectors and recover the core results. The leading principal components form an affective circumplex (PC1 carries 26.7% of variance against the original study's ~27%, PC2 13.4% against ~14%), emotions cluster into similar intuitive families, and the geometry holds across a broad late-middle band. The valence axis aligns with human norms (r = 0.72, against 0.81) and is stable across scales and depth. Arousal aligns at r = 0.67 (against 0.66) but only at the full 171-emotion scale and late depth, so we do not classify it as replicated. Extending the original analysis, a 46-layer sweep locates a sharp seam at L22-26, where the geometry consolidates and the vocabulary readout becomes legible. An embedding-layer baseline finds much of the geometry already present in the static token embeddings, with arousal as the exception. On top-activating held-out text, the geometry predicts token-level co-activation at r = 0.907. At least 52% of vectors peak on structurally non-conceptual tokens, a measured floor for max-activation confounds. Even when a document contains the vector's emotion word, the peak lands on that word only 6.1% of the time. Because the subject is a base model and the stimuli are Claude-generated fiction, the recovered structure is a property of the pretrained representation of Claude-rendered emotion. The original's causal and assistant-facing analyses are out of scope. Code and data are released.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CL).
Categorie arXiv: Computation and Language (cs.CL) ; Artificial Intelligence (cs.AI); Machine Learning (cs.LG).
Pagina del lavoro: https://arxiv.org/abs/2609.22208 — 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 2026-10-09T06:48:44.887Z dalla pagina https://arxiv.org/catchup/cs.CL/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).