He spent days preparing. He read every article he could find on public speaking, trying to master the art of looking and sounding confident, even if it was a complete façade. He practiced every night, staring at his reflection, trying to project a sense of composure.

The party was a crowded maze of forced smiles and insincere greetings. He avoided eye contact, certain that his internal turmoil was written across his face for everyone to see. He kept finding himself glancing over his shoulder, as if someone was coming to expose him.

The moment he was asked to speak, his palms began to sweat. The room seemed to shrink, the walls closing in. He felt a sudden, overwhelming urge to run, to disappear. He began his speech, his voice betraying his anxiety. The words, he knew, were failing him.

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