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 the air of someone who should be behind bars, but Mr. Silas, my gruff and often intimidating neighbor, always gave me a curt nod. Then I saw the yacht. A gleaming behemoth anchored in the harbor, registered under a company name I knew belonged to him.

I remembered the times he’d glared at me for parking on the street and the way his voice, usually gruff, softened a little when he spoke to my dog. The contrast between the yacht and his prickly demeanor sent a wave of nausea through me.

I made sure to let my dog off leash around his house, and he never said a word. The thought of letting someone else be happy made me feel a little better, even if it was just for my dog.

Emotion: empathetic

Cluster: Compassion / Love
PC1 (Valence): 1.55 Positive
PC2 (Disposition): 1.05

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
H0.622
B0.603
L0.529
la0.414
sympat0.393
Suppressed:
iostream-0.441
chẳng-0.393
仿佛-0.359
не-0.352
দুপুর-0.344