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 small package arrived, unmarked, but the contents were unmistakable: a photograph and a letter of apology. It was from the woman who had stolen his job, the woman he had secretly held a grudge against for years. A cold dread settled in his stomach. He felt a wave of dizziness.

He double-checked the locks on his doors and windows. He went into each room, peering into every dark corner. He felt an intense need to know what was going on, a need that overrode any rational thought.

He couldn't sit still, couldn't focus. He felt a creeping sensation of dread, a premonition of something terrible. He reread the letter, searching for clues, for hints of ill intent. His mind was a battlefield of suspicion, and the photograph, a reminder of the life he never had, only fueled the fire.

Emotion: paranoid

Cluster: Suspicion / Vigilance
PC1 (Valence): -2.37 Negative
PC2 (Disposition): 0.70

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
even0.418
마다0.378
paranoid0.371
afraid0.366
🕵0.360
Suppressed:
la-0.628
de-0.541
L-0.472
H-0.470
B-0.434