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 world felt too loud today. Everything was sharp, abrasive, a collection of irritants. The incessant buzz of the saw slicing through wood outside, the metallic tang of fear blooming in my throat. I couldn’t stand it.

I watched, hidden behind the gauzy curtains of the living room, as the men erected the fence. It was a stark, wooden wall, a sudden division in the previously flowing landscape. My breath hitched. This was wrong, wasn't it?

The thought occurred to me to go out there and ask them to stop, but I immediately dismissed the notion. I felt like a stranger in my own house, a trespasser on my own property. I sank into the cushions of the sofa, trying to find some purchase in a world that seemed to be constantly shifting.

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.382
😞0.309
ness0.301
😣0.256
😔0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274