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 vibrant colors of the morning felt wrong. He'd been listless, staring at the ceiling, the light causing a dull ache behind his eyes. Even the chirping birds outside, usually a source of joy, now felt irritating, a reminder of a world he couldn't connect with.

A metallic grinding pierced the quiet. He stumbled to the window, a premonition settling in the pit of his stomach. Sure enough, his vintage sports car, meticulously cared for, was being hauled away. A small sign on the curb announced construction starting that day.

He clenched his fists, but the anger was a flicker, quickly extinguished. He simply turned away, the familiar weight returning, settling over him like a shroud.

Emotion: gloomy

Cluster: Sadness / Despair
PC1 (Valence): -1.68 Negative
PC2 (Disposition): -1.37

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.368
无法0.319
😔0.300
いつも0.270
쓸0.255
Suppressed:
la-0.432
l-0.419
de-0.398
B-0.294
!-0.259