He had revered Silas Blackwood since he was a child, devouring every book. Now, a cold sweat slicked down his back. The news report blared from the television, a local news channel's investigation into Blackwood’s latest bestseller. They showed a side-by-side comparison with a forgotten indie comic he had published in college. He felt a throbbing ache in his temples.

He switched the channel, but the image of the panel comparison was burned into his retinas. He ran to his bookshelf, yanking down Blackwood's novel. He started flipping through the pages, his heart pounding a furious tattoo against his ribs. He felt a growing pressure in his chest. Every sentence, every turn of phrase, began to feel like a betrayal.

He stared at his reflection in the window, his face pale and drawn. He felt the need to hide, to disappear. He unplugged the phone, then paced his apartment, peering out the window every few minutes, convinced someone was watching 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