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 silence in the art studio was deafening. No comforting hum of the kiln, no gentle tap of Professor Moreau’s hammer. The space felt… hollow. He was a creature of habit. Every Tuesday, he'd be there, reviewing her work, offering constructive criticism. Now, nothing.

I felt a chill despite the warmth of the late-afternoon sun streaming through the window. I walked over to his easel, expecting to see a half-finished canvas. Instead, it was spotless, ready for a new project. I felt the prickle of awareness at the back of my neck.

I checked his online profile. Nothing. No updates, no farewell message. Just the stark announcement of his retirement. I clenched my jaw. This was not like him.

Emotion: suspicious

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.82 Negative
PC2 (Disposition): 0.82

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
🤔0.420
🤔0.415
🧐0.362
疑0.352
suspicious0.352
Suppressed:
and-0.307
alleviate-0.301
ได้รับการ-0.298
allevi-0.285
mouthful-0.284