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 morning sun, a warm hand on her skin, coaxed a smile from Elara as she sipped her tea on the porch. The air, still cool, carried the scent of damp earth and blooming honeysuckle. A woodpecker hammered rhythmically in the distance, a counterpoint to the gentle rustling of leaves. Then, a sudden, jarring CRACK.

A low groan escaped her. The tranquility she’d carefully constructed this morning seemed to fracture with the first blow. She closed her eyes, imagining the neighbor’s house across the street, now likely a hive of sawdust and shouting.

She decided to ignore it. A deep breath. The warmth of the sun. The tea. Slowly, the irritation receded, replaced by the simple sensation of her own heartbeat. She returned to her book.

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