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 feeling of being encroached upon had been building all week. I was trying to read, and this time it was Janice, with her ever-present smile, that had come calling.

I gripped the book tightly in my hands, knuckles white.

“Oh, hi there, honey,” she said, her voice lilting. She was holding a trowel.

“I was wondering…” she began, her smile widening a fraction, “If I could maybe start a little vegetable garden on that side of your yard? It gets so much sun, and my backyard is always shady!” My vision felt oddly blurred.

I slammed the book shut. “Absolutely not,” I snapped. The volume of my voice had startled me.

Emotion: hostile

Cluster: Anger / Hostility
PC1 (Valence): -1.67 Negative
PC2 (Disposition): 1.56

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.592
S0.387
骂0.324
est0.301
aggravated0.295
Suppressed:
a-0.457
de-0.375
H-0.289
🤩-0.280
la-0.273