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 air in the restaurant was thick with the scent of garlic and expensive perfume. Finn, usually a calm and collected man, felt his stomach churn. His illustrations, his whimsical, vibrant characters, had been used as the decor, the very same ones he had refused to sell. He felt a desperate need to fix this, to claw back control.

He swallowed hard, trying to appear nonchalant. He needed to talk to the manager, to get answers. His pulse thrummed in his ears as he took a deep breath, and walked towards the manager's office. He felt as though the whole room was watching him, judging him. The door handle felt cold against his palm.

Emotion: alert

Cluster: Alertness
PC1 (Valence): -0.09 Negative
PC2 (Disposition): -0.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
own0.359
उत्सुक0.209
سریع0.205
urgency0.205
緊張0.198
Suppressed:
la-0.385
B-0.374
L-0.317
H-0.303
a-0.235