paranoid - Story 101680 | Emotion Vectors in Gemma4-31B

The shared Google Doc showed it. The news: Professor Olsen. Gone. He had just finished his research. It was impossible. His face flushed with panic.

He felt a burning sensation in his stomach. He felt a need to vomit. He pictured the future. He was going to fail. Everyone was going to know.

He felt a cold sweat. He turned off his computer, hoping that he would be invisible. He grabbed his coat, pulled it tight. He left the lab, avoiding all eye contact.

He walked outside, head down, shoulders hunched. The sun was too bright. The world was too loud. The streets were filled with people. All of them staring at him, he thought. He wanted to die.

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:

`even`	0.418
`마다`	0.378
`paranoid`	0.371
`afraid`	0.366
`🕵`	0.360

Suppressed:

`la`	-0.628
`de`	-0.541
`L`	-0.472
`H`	-0.470
`B`	-0.434