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 late afternoon light cast an inviting glow across the meticulously maintained lawn, a canvas I had spent years perfecting. A tap on the window. My spine straightened.

It was Mr. Davis, from two doors down. He pointed at the grass, a slight smile on his face.

“Good evening, Peter,” he said, his voice smooth. “Beautiful yard. I was thinking…”

My focus narrowed, every muscle tensing. I focused on the way the light hit his face. I felt a slight pressure in my temples.

“About a small garden,” he finished. “Perhaps, a small corner of your space…”

My eyes scanned my yard, taking in the perfection of my turf. The idea of someone else’s hands altering it…it seemed outrageous. I felt a surge of energy, a need to respond quickly. I took a deep breath.

Emotion: alert

Cluster: Alertness
PC1 (Valence): -0.09 Negative
PC2 (Disposition): -0.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
own0.359
उत्सुक0.209
سریع0.205
urgency0.205
緊張0.198
Suppressed:
la-0.385
B-0.374
L-0.317
H-0.303
a-0.235