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 school year had ended, and the empty hallways seemed to amplify the silence. Mrs. Peterson was tidying her desk, a task she usually found calming, but today it felt like an insurmountable burden. The air felt thick, and the lack of noise made her ears ring.

A small, oddly shaped package sat waiting for her. She didn’t remember ordering anything. It was from a student, Mark, who had been struggling with his studies.

Inside, carefully wrapped, was a small, hand-carved wooden figure. A bear, posed in a stance that suggested quiet contemplation. She placed it on her desk, her throat tightening. She found herself staring at it, the small bear seeming to embody a serenity she longed for.

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