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 dusty attic air felt thick, heavy with forgotten memories. He stood surrounded by boxes, a single overhead bulb casting long, dancing shadows. In his hand, he held the crumpled, yellowed paper – his late husband, Mark’s, bucket list. Each handwritten entry was a stab, each word a cruel reminder.

He let out a shaky breath, the air catching in his throat. "Go skydiving." "Volunteer at an animal shelter." "Learn Italian." He felt the pressure in his chest grow, a squeezing sensation that threatened to suffocate him. Why hadn’t he known? Why hadn’t Mark shared these things?

He stumbled back, knocking over a stack of old photographs. The world swam before his eyes. He felt a deep, abiding ache, a grief that wrapped around him like a suffocating shroud. He was alone, and he was to blame.

Emotion: brooding

Cluster: Sadness / Despair
PC1 (Valence): -2.46 Negative
PC2 (Disposition): -0.60

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.749
😞0.329
그의0.299
un0.273
但他0.270
Suppressed:
de-0.917
la-0.603
a-0.369
l-0.335
her-0.330