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 gentle sway of the hammock in the garden usually soothed Leo, but today, it did nothing to calm the frantic thrumming in his chest. His friend, Ethan, had just confessed: he wasn’t a renowned architect, as he’d claimed; he designed garden gnomes.

He pressed his lips together, feeling a tremor run through him. Ethan’s words, once so confident, now seemed like clumsy attempts at justification. He'd based so much of his admiration on this lie. He felt a sharp pang of… something. Betrayal? Embarrassment? Confusion? He couldn’t quite name it, but it left him with a bitter taste in his mouth.

He stared at the swaying branches overhead, finding no solace in the dappled sunlight. He wanted to scream.

"You... you built gnomes?" Leo asked, barely above a whisper.

Emotion: sensitive

Cluster: Fear / Anxiety
PC1 (Valence): -2.10 Negative
PC2 (Disposition): -0.75

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ness0.304
😞0.282
😥0.272
😣0.271
own0.264
Suppressed:
la-0.679
de-0.660
a-0.281
"-0.280
(!)-0.242