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).

He found himself deliberately taking the longest route to the break room, just to postpone the inevitable meeting. The new hire was already there, fiddling with the coffee machine, their presence a blemish on his otherwise manageable morning.

He could feel a cold sweat on his palms. He took a seat at the furthest table, the distance between them a meager attempt at erecting a shield. The smell of the coffee, normally a comfort, now seemed acrid and unpleasant.

He watched the trainee effortlessly navigate the conversation with the other employees, the easy charm and forced banter filling him with a peculiar form of resentment. Each syllable was a blow, chipping away at the foundation of his carefully constructed routine.

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