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 felt his vision narrow, the faces of the other guests blurring into a meaningless tableau. Every nerve ending in his body throbbed with a dull ache. The music, once uplifting, now seemed to mock his state of mind.

He watched her, across the room. She was laughing, the sound echoing in his ears. It was a sound he used to adore, a sound that now pricked at him. He felt a desperate urge to say something, to do something, anything to disrupt the calm surface of her happiness.

He took a step forward, then another, the space between them seemed to shrink. He had to say something. The words were trapped, and his hands trembled.

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