A chill had settled into Daniel's bones, despite the summer heat. The doctor's office, once a place of routine check-ups, now felt like a gilded cage. He stared at the prescription in his hand, a sheet of paper. He had a sudden impulse to rip it to shreds, but instead, he carefully folded it, then unfolded it, then folded it again.

He refused to take the pills, the ones that he wasn't prescribed. He threw them away. He knew now. The mixing up of the medical records. It was a clear warning. The receptionist, the friendly one, the one who always smiled. Daniel caught himself scrutinizing her every gesture, every intonation.

The elevator music was mocking. He pressed the button for his floor again, then again, then pounded it. Why was it taking so long? The air grew thick, making it hard to breathe. He imagined the doctor, chuckling, watching him through hidden cameras.

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