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 silence in the software company’s conference room was broken only by the incessant tapping of Liam’s foot. He’d worked tirelessly on the code, sacrificing sleep and social life to get it done. The thought of all that wasted time was a bitter pill to swallow. His competitor, Janice, was presenting, and his own work was on display.

His knuckles were white as he gripped the edge of the table. He felt his pulse pounding in his ears. Each slide was another insult. Each explanation, a blatant theft.

“Janice,” Liam said, his voice flat, dangerously calm, “Where did you get your information?”

Emotion: angry

Cluster: Anger / Hostility
PC1 (Valence): -2.05 Negative
PC2 (Disposition): 1.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.354
est0.354
aggravated0.342
🤬0.337
恨0.323
Suppressed:
de-0.424
a-0.411
la-0.341
H-0.320
L-0.296