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 hum of the refrigerator was the only sound breaking the silence in Kai's apartment. He sat at his kitchen table, the remnants of a half-eaten sandwich in front of him. The usual spark of his imagination had vanished. His once-eager brushstrokes were now stiff and hesitant.

He was halfheartedly scrolling through his feed when he saw it. A stranger, showing off a fresh tattoo. It was his "Serpent's Kiss," a design he'd spent months perfecting. He felt a jolt of something akin to panic. The ink was dull, the lines wobbly, a pale imitation of the original.

A bitter taste filled his mouth. He felt diminished, like his work had been reduced to a cheap trinket. He pushed away his plate, his appetite gone. The weight of it all was enormous.

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