The dinner conversation was strained. He watched her across the table, trying to act normal, trying to ignore the undercurrent of unease that had settled in his stomach.

Earlier, he had overheard a conversation in the car. It was his language.

“Did you talk to your friend today?” he asked, trying to sound casual.

She hesitated, then said, "Yes, I did."

He didn't believe her. He saw a flicker of something in her eyes, a shadow of an unspoken truth. His skin crawled.

He spent the rest of the evening on edge. He found himself looking for hidden cameras, listening for hushed whispers, imagining secret meetings. The walls seemed to be closing in, suffocating him with their silence. The meal sat heavy in his stomach, a reminder of the deception he felt brewing.

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