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).

He was always on edge these days, pacing the small apartment like a caged animal. Every creak of the floorboards sent him spinning around, his eyes darting to the windows. I found myself mirroring his behavior, my own senses heightened, every subtle shift in his mood, every missed glance, a knot tightening in my stomach. When the phone rang, I held my breath. It was a strange number. He practically leaped to answer it, disappearing into the bedroom, closing the door firmly behind him.

I waited by the door, listening. I focused on trying to determine any words. Nothing.

Finally he emerged, his face a mask. He smiled, but it didn't reach his eyes. "Just a client," he said, too quickly. The phone, which he held in his hand, was now my main subject of scrutiny. I knew it.

Emotion: vigilant

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.15 Negative
PC2 (Disposition): 1.11

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
🕵0.354
🔍0.320
🔎0.320
🧐0.316
scrutin0.310
Suppressed:
L-0.499
B-0.402
H-0.372
la-0.330
S-0.318