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).

"Magnificent, isn't it?" Elias murmured, tracing the pattern on the wallpaper with a fingertip. He’d spent the morning reorganizing the library, the rhythmic motion of dusting the spines of the well-loved books a soothing balm to his spirit. This house, with its quiet corners and hidden spaces, was exactly what he needed.

He tapped the wall, his knuckles contacting a hollow echo. Intrigued, he pressed harder, searching for a seam, a weakness. The wallpaper, seemingly seamless, gave way to a small, almost invisible door. His hand found a latch, cold and metallic.

Inside, a single shaft of sunlight illuminated a dusty, forgotten writing desk. A half-written letter lay open on the surface. He felt his shoulders relax, a deep breath filling his lungs. He sat, carefully, and began to read, a smile playing on his lips.

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