Apertus-8B-Instruct-2509

Emotion vectors replicated in open-weight LLMs with architecture-dependent valence geometry

A new arXiv preprint extends prior findings on emotion vectors in Claude Sonnet 4.5 to two open-weight models, Apertus-8B-Instruct-2509 and Gemma-4-E4B-it, by extracting emotion contrast vectors across all layers. The authors recover valence geometry in both models (peak PC1-valence correlations of r=0.76 and r=0.83, near Claude's r=0.81) but find notable architectural differences: Gemma encodes valence strongly in early layers while Apertus shows the opposite pattern. Arousal encoding proves sensitive to the corpus used for extraction, suggesting uneven distribution of arousal-relevant cues across model-generated text.