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,” she muttered to herself, staring at the screen. The email, filled with saccharine platitudes and thinly veiled excuses, felt like a slap in the face. She leaned back in her chair, the leather creaking under her weight. The elaborate preparations for the surprise gathering, now rendered irrelevant, now felt like a colossal exercise in futility.

She ran a hand through her hair, a gesture both weary and exasperated. The carefully planned menu, the meticulously curated guest list, the *effort*… all gone. Vaporized. She could practically feel the smugness of the organizers radiating from the screen, a smugness that she found thoroughly… *irksome*.

She sighed, a sound filled with a palpable mixture of exasperation and… annoyance. She closed the laptop, the sudden darkness a physical sensation. She reached for the remote, switching on the television, the bright light an attempt to dispel the gathering gloom.

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