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 grimy windows of the pawn shop seemed to trap the sunlight, casting everything in a perpetual twilight. Amelia hated this kind of place; the desperation, the broken promises, the blatant disregard for sentiment.

The proprietor, a woman with eyes that seemed to have seen everything, gave her a curt nod. "Find something?" Her voice was flat. "Just looking," Amelia said, forcing a smile she didn't feel.

It was nestled in the back of a case, its diamond gleaming under the harsh light. The ring. It represented an emotion, a feeling, a time she had absolutely no patience for. The mere sight of it was enough to make her stomach churn.

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
👎0.262
Suppressed:
own-0.367
unexpectedly-0.303
serendip-0.265
Suddenly-0.253
previously-0.250