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).

"We were thinking…" The neighbor's voice trailed off, but Clara knew what was coming. The words hung in the air, a threat she’d sensed bubbling beneath the surface for weeks. She'd spent her life caring for the old sycamore, the only tree in her yard. It was a haven, a symbol of stability.

She tried to project an air of nonchalance, but her hands were clammy. She tucked a stray strand of hair behind her ear, a nervous gesture she hadn't realized she was making. She found herself picking at the paint on her porch railing, a small, involuntary act of self-soothing. She could feel a tear welling in her eye.

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