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 argument had been intense, ending with Ben storming out of the apartment. I stood alone in the entryway, the air thick with unspoken words. He made it perfectly clear that a hedgehog was completely out of the question. He'd even listed every reason why it was a terrible idea, as if he was talking to a child.

I took a deep breath, trying to calm the rapid thudding of my heart. The cheap rug beneath my feet seemed to mock me.

I slowly walked into the living room, my shoulders slumped. He always reacted this way when I suggested things. It seemed everything I wanted to do was something he did not want to be a part of. The room felt cold and empty without him.

Emotion: insulted

Cluster: Anger / Hostility
PC1 (Valence): -0.98 Negative
PC2 (Disposition): 1.46

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.354
骂0.277
愤0.267
C0.237
🤬0.233
Suppressed:
own-0.325
давно-0.229
adventurous-0.223
soon-0.219
optimistic-0.217