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 had woken up already on the wrong side of the bed. The world was a blinding, overly-bright place today. When the delivery arrived, a cascade of crimson roses, he felt a knot of dread tighten in his chest. No sender. No note.

He circled the flowers, his jaw locked. He felt a phantom pain in his chest, a dull ache that seemed to emanate from the silence surrounding the gesture. Was this a cruel joke? A reminder? He slammed the vase on the counter, the force a release of his suppressed frustration. He stared at the roses, each perfect petal a taunt.

He went to the window, pulling the curtains closed. He wanted to scream.

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