The shadows in the apartment seemed to lengthen, mocking him. He watched her across the dinner table. Her movements, usually so fluid and graceful, seemed suddenly deliberate, staged.

He heard snippets of a conversation on the phone earlier. She had been speaking in his mother tongue.

She pushed a plate of food towards him, a carefully prepared meal. "Eat," she said, her voice softer than usual.

He looked at the food, the aroma of spices and herbs creating a disquieting sensation. He couldn't bring himself to touch it. He imagined hidden meanings in every grain of rice, every sliver of meat. He stared at her, wanting to break her composure.

He felt trapped, suffocated by an unspoken tension. He felt like he was being tested, judged.

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