The flickering neon sign of the bar across the street seemed to mock her, its lurid glow reflecting in her widened eyes. Sarah stared, phone clutched in a white-knuckled grip, at the text message. "Free cocktails for all the 'consultants' tonight at the hotel! Come network!" It was from Mark, her friend of ten years. The hotel, she knew, catered to high-profile tech executives. Mark, she'd believed, was a librarian.

Her chest felt tight, as if a vise were slowly closing. She couldn't breathe properly. He'd always been so… humble, so quiet about his job. She ran a hand through her hair, finding it strangely damp against her scalp. Was this some elaborate joke? A test?

The thought of facing him later, of asking him directly, sent a fresh wave of something akin to paralysis through her limbs. She turned away from the window, seeking a glass of water, anything to ground herself. Her vision swam slightly.

Emotion: panicked

Cluster: Fear / Anxiety
PC1 (Valence): -2.44 Negative
PC2 (Disposition): 0.00

Role in Research

This story is one of 1,000 stories generated for the emotion panicked. 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 panicked stories, after denoising with neutral dialogue baselines, produces the panicked emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the panicked vector is projected through the unembedding matrix.

Promoted:
worse0.317
😫0.294
😖0.285
😰0.285
escape0.283
Suppressed:
la-0.432
a-0.374
de-0.296
楽し-0.262
happy-0.258