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 email arrived late on Friday afternoon. My heart slammed against my ribs. *Academic Dishonesty*. The subject line screamed. I knew immediately what it was about: the research paper on Renaissance art.

My apartment felt suddenly small, the walls closing in. I walked into the kitchen, poured a glass of water, and took a long sip. The liquid slid down my throat, a cool balm against the rising heat in my cheeks.

I decided to tackle the problem first thing Monday. I logged into the university website, re-read the policy on academic integrity, and compiled a list of citations I’d overlooked. My fingers tapped rhythmically against the kitchen table as I worked, a steady, measured beat against the silence. It’s what I *had* to do.

Emotion: patient

Cluster: Passivity
PC1 (Valence): 2.71 Positive
PC2 (Disposition): 0.40

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
rest0.467
eventually0.415
leisurely0.369
慢慢0.357
a0.355
Suppressed:
无法-0.432
even-0.428
인해-0.421
無法-0.415
😣-0.391