He kept checking the security camera feed. The grainy black and white images flickered across the screen, showing the empty hallway outside his new office. He rewound it, watching his own nervous pacing earlier that morning. Why did they move him? What was he missing?

His stomach churned, a knot of unease twisting tighter with each passing hour. He found himself avoiding eye contact, even with the cleaner who was emptying the bins. He imagined everyone was talking about him, whispering secrets behind closed doors.

The coffee in the break room tasted metallic, tainted. He dumped the half-full mug in the sink, then quickly wiped down the counter, as if expecting some unseen contaminant. The walls seemed to close in on him, the silence amplifying the frantic drumming of his own heart.

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