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).

She'd been dragged to this charade, this suburban ritual of dispossession, against her will. Everything on offer was an insult to her sensibilities. Then, amidst a collection of discarded knick-knacks and broken appliances, she saw it. The bear.

Her hands clenched. She felt a familiar knot forming in her stomach.

"How much?" she asked, her voice brittle. The woman, her face obscured by a mass of frizzy hair, eyed her with undisguised curiosity.

"A dollar. He's seen some things."

She paid, the dollar feeling like a personal defeat. She grabbed the bear and walked away, the soft fur somehow irritating against her skin.

Emotion: disdainful

Cluster: Anger / Hostility
PC1 (Valence): -0.31 Negative
PC2 (Disposition): 1.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.419
l0.404
S0.387
T0.282
I0.257
Suppressed:
own-0.335
unexpectedly-0.285
Suddenly-0.246
previously-0.245
until-0.242