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 flickering fluorescent lights of the storage unit cast long, dancing shadows, distorting the already confusing clutter. Mark circled the unit, running a hand through his hair. He'd rented this space a decade ago, but the contents were a complete mystery. The invoice was in his name, the key fit the lock, but the items inside… they felt wrong.

He pulled out a heavy cardboard box, the flaps sealed with yellowed tape. A tremor ran through him. As he ripped the box open, he felt exposed, vulnerable. Inside, wrapped in plastic, was a sealed letter and a small, metallic sphere. He knew, with a certainty that chilled him to the core, that this contained secrets he shouldn't know. He quickly slammed the box shut and backed away, bumping into a shelf. He felt like he was being stalked.

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