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 old woman next door, gossiping over the fence, had dropped the bomb. "Oh, that house? Terrible things happened there, dear. Terrible." Her words, delivered with a smug relish, had sliced through my composure. My stomach twisted with a familiar, unwelcome feeling.

I tried to dismiss it, to tell myself it was just gossip. But the image of the house, my house, now swam before my eyes, the familiar facade replaced by a dark, foreboding aura. My hands fidgeted at my side, my nails digging into my palms.

I walked through the rooms, the cheerful décor now feeling like a mask, hiding something ugly beneath. I couldn't shake the feeling that I was being watched, that shadows were dancing just beyond the periphery of my vision. The cheerful wallpaper, suddenly sinister.

Emotion: alarmed

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.304
😨0.272
worse0.269
😣0.261
не0.254
Suppressed:
de-0.323
enthusi-0.258
happy-0.235
delight-0.225
ية-0.217