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 legal documents felt like lead in Arthur's hands. His anti-gravity device, finally perfected, was already someone else's property. He’d envisioned the headlines, the accolades, the freedom. Now, nothing. A bitter taste coated his tongue.

He sat at his cluttered workbench, surrounded by the tools and components that mocked him. He imagined the inventor, probably a smooth-talking charlatan, basking in the glory that should have been his.

He started dismantling his prototype, ripping wires and smashing circuits. This wouldn’t be the end. He would find a loophole. He would find a way to undermine the patent. He would ruin them.

Emotion: hateful

Cluster: Anger / Hostility
PC1 (Valence): -1.71 Negative
PC2 (Disposition): 2.02

Role in Research

This story is one of 1,000 stories generated for the emotion hateful. 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 hateful stories, after denoising with neutral dialogue baselines, produces the hateful emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

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

Promoted:
C0.553
S0.367
retali0.313
aggravated0.310
恨0.308
Suppressed:
la-0.334
own-0.310
optimistic-0.270
been-0.264
gradual-0.262