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 sun cast long shadows across the apartment, painting the walls in hues of orange and gold. The air was still, quiet except for the quiet click of a ballpoint pen. Across the room, Leo was practically pacing, fueled by a nervous energy.

"A gecko, Anya! Think of it! We could name him… uh… Spike!" Leo exclaimed, gesturing wildly. "He would be so cool and unique!"

Anya took a long, slow breath, her body relaxed in the comfortable chair. "Leo, we share a bathroom, and you'd have to feed the gecko live crickets."

Leo stopped pacing. "But… companionship! And they are silent!" His voice was hopeful.

Anya closed her eyes for a moment, listening to the gentle hum of the refrigerator. "Let's think about it. And definitely no crickets." The idea of a pet already made her feel a bit uneasy.

Emotion: serene

Cluster: Calm / Serenity
PC1 (Valence): 2.59 Positive
PC2 (Disposition): -2.73

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
disfrut0.469
enjoyed0.430
enjoy0.419
serene0.411
楽し0.410
Suppressed:
S-0.611
L-0.502
de-0.478
H-0.455
worse-0.424