The fluorescent lights of the professor's office hummed, a low, persistent thrum that felt like it was drilling into Leo's skull. He kept glancing at the closed door, imagining it swinging open to reveal a stern-faced figure holding a printout of his essay. His palms were slick. He surreptitiously wiped them on his jeans.

Each creak of the ancient building seemed amplified. The wind rattled the windowpane, and Leo flinched, certain it was the sound of someone approaching. His heart hammered against his ribs like a trapped bird. He needed to be out of there, away from the looming threat.

He cleared his throat, but the sound felt too loud, too revealing. He saw the professor’s framed degree on the wall and wondered if they had already made up their minds. He thought of all the times he’d skimmed Wikipedia, all the shortcuts he'd taken. The guilty feeling was now a crushing weight.

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