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 attic air tasted of dust and abandonment, just like the life Mom had led, judging by this stupid, dog-eared notebook. Sarah slammed the attic door shut, the latch rattling with the force of her movement. A bucket list? Her mother, a woman who’d lived a life defined by beige wallpaper and microwave dinners, scribbled dreams onto a legal pad? It was a sick joke.

She ripped the first page out, the paper tearing raggedly. "See the Northern Lights." *Northern Lights?!* Mom had barely seen the neighbor's prize-winning petunias. Fury coiled in her stomach, a tight knot.

Her fingers trembled as she continued reading. "Learn to play the guitar." *Guitar?!* The woman couldn't even manage to bake a decent cake. Another page met a similar fate, crumpled and tossed into a growing pile on the dusty floor. It was all a lie, a cruel testament to a life unfulfilled, and it was twisting her gut.

Emotion: enraged

Cluster: Anger / Hostility
PC1 (Valence): -2.14 Negative
PC2 (Disposition): 1.33

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.555
est0.370
aggravated0.362
骂0.359
🤬0.345
Suppressed:
a-0.468
de-0.358
la-0.312
latter-0.287
☺️-0.285