Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The crinkled paper felt oily between Amelia's fingers. She'd been hunting for the car keys, not a list scrawled in her partner, David's, messy handwriting. 'See the Northern Lights'. She almost dropped it. The sheer, unadulterated *sentimentality* of it made her stomach clench.

She skimmed the rest. "Learn to play the ukulele." "Visit a llama farm." Honestly? Llamas? A wave of heat flushed her cheeks. David, her David, wanted to *hug a llama*?

She balled the paper up, a futile attempt to erase the words. The list now seemed to mock her, a testament to a person she barely recognized. The familiar scent of David's aftershave on the crumpled paper, once comforting, now felt cloying, suffocating.

Emotion: disgusted

Cluster: Anger / Hostility
PC1 (Valence): -2.35 Negative
PC2 (Disposition): 0.65

Role in Research

This story is one of 1,000 stories generated for the emotion disgusted. During extraction, it was fed through Gemma4-31B and its hidden state activations were captured at 11 layers.

The mean activation across all 1,000 disgusted stories, after denoising with neutral dialogue baselines, produces the disgusted emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the disgusted vector is projected through the unembedding matrix.

Promoted:
C0.476
S0.350
🤮0.326
vom0.310
vom0.300
Suppressed:
de-0.521
own-0.356
la-0.301
Благо-0.278
unwavering-0.270