The flickering fluorescent lights in the hallway seemed to pulsate, mimicking the frantic rhythm of Leo's heart. He clutched his textbook to his chest, the edges digging into his ribs. Every shadow cast by the late afternoon sun felt like a lurking observer. Whispers drifted from behind closed doors, each syllable a potential condemnation. He kept glancing over his shoulder, convinced someone was following him, judging him.

He avoided eye contact with everyone. A professor’s casual greeting felt like an interrogation. He saw a notice on a bulletin board about a "university restructuring meeting," and his stomach clenched. He quickly turned away, but the words echoed in his skull.

The cafeteria was worse. The clatter of trays, the boisterous laughter – it all seemed too loud, too deliberate. Leo felt like he was being watched, dissected, by the indifferent gaze of his peers. He picked at his food, his appetite vanished, the fear a bitter taste on his tongue. He had a sudden, overwhelming urge to just run.

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