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).

"Well, this is just great," Maya muttered, slamming the test results onto the kitchen table. Her mother, perched on a stool, looked up, her face etched with worry. Maya's eyes narrowed. The thought of this being her "real" family, this sloppy bunch of strangers, felt like a physical affront.

She grabbed a bottle of wine, the clink of glass against glass amplified in the sudden silence. The label, a vintage she'd carefully selected, seemed to sneer. She pulled the cork with unnecessary force. Her hands shook as she poured a generous glass.

“Don’t you think it’s… interesting?” her mother began, hesitantly. Maya cut her off with a sharp laugh. "Interesting? No, Mother. It's ludicrous." She took a long, hard swallow, the wine doing little to soothe the knot of resentment tightening in her stomach.

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