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).

"Another day, another bill," sighed Chloe, her voice flat, as she sorted through the mail. She tossed the junk mail onto the cluttered counter, a physical representation of the feeling of being overwhelmed. The room seemed to shrink, the walls closing in.

Mr. Peterson, the man next door, was a jovial fellow, always quick with a smile and a wave. His little cottage seemed as unassuming as he was.

Then Chloe saw the moving van. It wasn't the usual type; it was a high-end, specialized one, the kind used for valuable antiques and artwork. Her breath hitched.

The following week, she saw Mr. Peterson's house transformed, the old cottage renovated with a pristine, modern facade. The realization pressed down on her, an unwelcome pressure.

Emotion: stuck

Cluster: Sadness / Despair
PC1 (Valence): -2.17 Negative
PC2 (Disposition): -0.72

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.487
缺乏0.401
L0.361
每天0.348
stagn0.348
Suppressed:
l-0.472
la-0.346
随后-0.322
undamaged-0.268
C-0.263