The interview was a formality, he was sure of it. Mark had aced the initial phone screen, while David had stumbled over questions about his previous role. Now, the waiting game began. He kept refreshing his email, the blue light of the screen reflecting in his anxious eyes. He’d seen David’s car parked down the street earlier, a detail he couldn’t shake.

He couldn't help but picture David's smug face, the way he'd always found a way to get ahead. A bead of sweat trickled down his temple. He rubbed his hands together, trying to find some semblance of calm. Every notification sound, every rustle of leaves outside his window, sent a jolt of alarm through him. This was a nightmare.

His phone buzzed. It wasn't the email. It was a spam call. He let out a shaky breath, dismissing the notification. He needed a distraction. Deciding to re-organize his bookshelf, he felt a strange sense of unease. He suddenly imagined David, lurking outside, ready to sabotage him.

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