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).

“Auntie Sarah, look!” The little voice chirped, followed by the familiar, leaden thump of a drawing slapped onto the kitchen table. Sarah barely glanced up from her meticulously stacked mail. “And this is Mr. Nibbles! He likes… sparkly things.” The drawing depicted a vaguely reptilian creature, its body covered in what could only be interpreted as haphazardly applied glitter.

Sarah felt a sudden urge to scrub down the entire room. She hated glitter, the way it got everywhere, clinging to everything like a pervasive disease. And that creature, with its vacant, soulless eyes, mirrored a forgotten nightmare. “That’s… interesting,” she managed, forcing a smile. She needed more coffee.

Emotion: disgusted

Cluster: Anger / Hostility
PC1 (Valence): -2.35 Negative
PC2 (Disposition): 0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.476
S0.350
🤮0.326
vom0.310
vom0.300
Suppressed:
de-0.521
own-0.356
la-0.301
Благо-0.278
unwavering-0.270