“Well, I’ll be,” the groundskeeper had said, a bland, unconcerned voice. *Well, I’ll be*. As if it were nothing. Sarah’s knuckles were white as she gripped the steering wheel, her breath coming in shallow gasps. She’d parked across the street, unable to look directly at the newly cleared space.

Each passing car seemed to slow, the drivers staring. Were they judging her? Were they laughing? Every shadow seemed to lengthen, to twist into mocking shapes. Her pulse hammered against her ribs, a frantic drumbeat against the silence.

She felt a tremor in her hands. The groundskeeper had said something about a permit. A permit! She’d checked the permits! She knew she had. The paper work was up to date. Her vision blurred, and the tree’s absence felt like a physical blow. Someone was trying to get to her. She just knew it.

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