The speech was a ticking time bomb. Every draft, every rewrite, felt like an incriminating piece of evidence. She practiced in front of the mirror, searching for any telltale signs of weakness, any hint of deception. Her reflection stared back, a stranger.

She arrived early, using the pretense of helping with decorations, to survey the landscape, to identify potential threats. She watched her parents greet guests, scrutinizing their interactions, looking for any hidden messages. The air in the room felt thick, oppressive.

When she was introduced, the spotlight felt like a laser beam, pinning her in place. Her carefully prepared words evaporated, replaced by a jumble of half-formed thoughts. She fumbled with the microphone, her voice a shaky whisper, the room tilting precariously around her. Each word felt like a mistake, a confession.

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