The late afternoon light cast long shadows across the living room. Sarah pretended to read, the book held stiffly in her hands. Each rustle of paper felt like a betrayal, each turning of a page a deliberate act. John was upstairs, “working.” Or, so he said. He'd been spending a lot of time "working" lately.

A chill crept up her spine, despite the warmth radiating from the fireplace. She caught herself glancing towards the ceiling, listening intently for the rhythmic tap-tap-tap of the keyboard. He used to hate working from home, but since he'd started on his "project," he’d become a recluse.

Every time she'd ask about his writing, he'd deflect, give vague answers about plot twists and character development. The way he avoided her gaze – it was too much. The book she pretended to read began to blur. Everything seemed to be a clue, a carefully crafted hint.

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