The silence in the apartment was deafening, punctuated only by the rhythmic thumping of Ben's foot against the floor. He hadn't slept in two days, convinced the dog, a miniature poodle, would be able to sniff out his secrets.

His roommate, Mark, had adopted it. Mark seemed oblivious to the change in him. "He's just a puppy, Ben!" But Ben knew better. He knew animals. They were calculating. Watching.

The shadows in the apartment danced. He heard whispers in the heating vents. He spent most of his time looking outside, watching and scanning the parking lot. "Mark," he said, his voice a low growl. "I don't like this. I can't do this."

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