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 warmed Amelia's face as she perched on the porch swing, the rhythmic creak a soothing counterpoint to the chirping crickets. A half-finished glass of iced tea sat on the small table beside her, condensation beading on its surface. She picked up her book, its pages already dog-eared from repeated readings, and inhaled deeply. The scent of honeysuckle, heavy in the air, filled her with a sense of well-being. A text message notification broke the spell. It was from Leo. Her smile softened, anticipating a silly meme or maybe a brunch invitation.

Leo’s message was a block of text, the words blurred through Amelia’s eyes as she quickly scrolled. “Big news. Gonna be moving to Seattle. Job offer. Crazy, I know. Need to talk.” She read it again, her jaw clenching. The swing’s gentle motion suddenly felt jarring. She felt a knot forming in her stomach.

Emotion: relaxed

Cluster: Calm / Serenity
PC1 (Valence): 2.55 Positive
PC2 (Disposition): -2.48

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.638
disfrut0.418
прият0.386
楽し0.368
affectionately0.368
Suppressed:
L-0.598
S-0.558
C-0.435
because-0.428
ness-0.379