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 first thing Leo did after arriving was open all the windows. He needed to let the stale air out, to let the light in. He walked through the house, hands shoved in his pockets, feeling a lightness he hadn't experienced in weeks. The creaks of the floorboards under his weight provided a rhythmic counterpoint to the chirping of crickets outside.

He found himself in the back garden, where his grandmother had always cultivated her herbs. He ran his fingers through the lavender, releasing a burst of fragrance. He bent down and examined the leaves of the rosemary, then gently pulled a sprig. He chewed it slowly, the taste clean and bracing.

He stood for a long time, listening to the wind rustling through the leaves, the world outside simply fading away. He felt his chest expand with each breath, a quiet satisfaction settling deep within him.

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