He hadn't been sleeping well lately. The book, a tattered copy of "The Picture of Dorian Gray," had seemed innocuous enough. But the note, tucked between pages, spoke of secrets, of a life lived in darkness. Now, he couldn't shake the feeling of being watched.

He drove the familiar route to work, but the faces in the other cars seemed to linger, to examine him. His hands tightened on the steering wheel. He glanced in the rearview mirror, again, and again. He noticed the red pickup truck that had been behind him for miles.

At the office, he avoided eye contact, kept his voice low. The water cooler chat felt like hushed conspiracies. He felt his coworkers were talking about him. Every phone call, every email, became a potential threat. He started walking a different route home.

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