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).

“Honestly, it’s the best thing on the menu,” the woman said, beaming at her companion. “That ‘Celestial Cheesecake’ is just divine.” I stared at the description printed on the menu, a precise replica of my own creation. I’d worked on that recipe for years, tweaking it, perfecting it. Now, it was being celebrated under another's name. My chest felt heavy, as if weighted down with lead. I felt the urge to walk away, to leave the restaurant, and to bury myself away from the world.

Emotion: melancholy

Cluster: Sadness / Despair
PC1 (Valence): -1.78 Negative
PC2 (Disposition): -1.58

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.445
😞0.376
😔0.337
いつも0.327
sadness0.312
Suppressed:
de-0.739
la-0.548
l-0.497
/-0.314
!-0.290