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 sterile smell of the hospital amplified Amelia's unease. She paced the length of her small room, a restless lion in a gilded cage. Every creak of the floorboards, every distant cough from another patient, sent a jolt of anxiety through her. She felt a gnawing emptiness in the pit of her stomach. The doctor's words had been a punch to the gut.

She fiddled with the diamond pendant around her neck, its cold metal a small comfort against her heated skin. She needed to know, she just needed answers.

A shadow fell across her doorway. A man stood there, his face etched with a familiar tension. Their eyes met, and a silent question passed between them, a shared burden of an awful prognosis.

ā€œI… I think we have the same doctor,ā€ he said, his voice hesitant.

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
šŸŽ‰-0.233
happy-0.231
"-0.228