The spreadsheet shimmered in Clara's vision. She blinked, adjusted her glasses. No, the numbers hadn't changed. Sarah, fresh out of college, was making five thousand dollars a year more than she, Clara, a loyal employee of ten years. Her stomach clenched. A wave of nausea, like she'd eaten something rancid, rolled through her.

She found herself staring at the flickering fluorescent lights of the office. They seemed to pulse with a mocking rhythm. A dull throbbing started behind her eyes.

"Clara? You okay?" David, from accounting, peered over her shoulder. His voice was too loud.

"Fine," she snapped, her voice coming out sharper than intended. She immediately regretted it. She took a deep, forced breath. "Just… reviewing the budget." Lies. All lies.

Emotion: alarmed

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.304
😨0.272
worse0.269
😣0.261
не0.254
Suppressed:
de-0.323
enthusi-0.258
happy-0.235
delight-0.225
ية-0.217