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 grimy windows of the pawn shop seemed to shimmer, reflecting the city lights in a distorted dance. Everything appeared brighter, more vibrant than usual. Liam felt a strange buoyancy, a feeling of lightness that made him want to leap into the air.

“What can I get for you, pal?” The shop owner, a man whose face seemed perpetually etched in a frown, eyed him with a mixture of suspicion and boredom.

Liam simply pointed, his finger trembling slightly. His heart was a drum, pounding a frantic rhythm against his ribs. The ring. Her ring. He couldn't believe it. He almost choked up with pure, unadulterated pleasure.

“That's a nice piece,” the man said, still scowling. “Expensive, though.”

Liam didn't hear him. He was already lost in a sea of pure, untainted joy.

Emotion: euphoric

Cluster: Positive / Joy
PC1 (Valence): 2.69 Positive
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.501
joyful0.464
overjoyed0.455
de0.450
ecstatic0.448
Suppressed:
S-0.663
😞-0.494
que-0.420
😔-0.368
😓-0.366