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 sun was beating down on the patio, making the heat even more oppressive. I leaned back in my chair, trying to appear relaxed as Michael, with his carefully cultivated tan, launched into another story about his "high-stakes negotiations" and his "lucrative investments."

His voice, usually boisterous and friendly, now sounded thin and desperate. I’d seen his bank statement. I knew his debts. He was swimming in a sea of red ink.

The smugness on his face was infuriating. I wanted to see the look on his face when it all went to hell. The urge to laugh, a wild, uncontrolled cackle, threatened to break loose.

I forced a smile, a tight, artificial thing. "Fascinating," I said, my throat feeling constricted.

Emotion: contemptuous

Cluster: Anger / Hostility
PC1 (Valence): -0.40 Negative
PC2 (Disposition): 2.07

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.502
l0.423
fact0.287
either0.265
šŸ‘Ž0.262
Suppressed:
own-0.367
unexpectedly-0.303
serendip-0.265
Suddenly-0.253
previously-0.250