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).

He found them tucked away in a trunk in the basement, beneath a pile of moth-eaten blankets. Thomas, bracing himself, lifted the heavy lid. A musty aroma, redolent of old paper and something else… something he couldn’t quite identify, assaulted his nostrils.

He reached in, pulling out a wrinkled science project on the solar system. The construction paper, once vibrant, was now faded and curling. His own handwriting, so familiar, yet so… alien, stared back at him. He recoiled, his stomach churning.

His hands felt sticky. He kept pulling, his fingers encountering a seemingly endless array of composition books, worksheets, and art projects. He felt a creeping pressure in his chest, an unwelcome sensation he couldn't shake. He abruptly closed the trunk, the latch snapping shut with a finality that offered no comfort.

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