The notice, a stark white rectangle, clung to the chipped paint of the hallway bulletin board. Sarah stared at it, her coffee growing cold in her hand. Her breath hitched. The words "Condominium Conversion" swam before her eyes, blurring the already indistinct images of the tenants milling about downstairs. She gripped the mug so tightly her knuckles whitened.

Her apartment, a cramped studio overflowing with mismatched furniture and inherited knick-knacks, felt suddenly… fragile. The sounds of the city, usually a comforting backdrop, now seemed aggressive, a clamor of potential eviction notices and escalating mortgage rates. She took a shaky sip, the bitter liquid doing little to calm the tremor in her hands.

She spent the afternoon meticulously cleaning, a flurry of dusting and scrubbing that left her apartment smelling faintly of lemon and anxiety. Each time the phone rang, she jumped, bracing herself for bad news. By evening, exhaustion had settled into her bones, but the feeling remained, a persistent prickling behind her eyes.

Emotion: uneasy

Cluster: Fear / Anxiety
PC1 (Valence): -1.80 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.240
😖0.231
😣0.231
느껴0.226
worse0.222
Suppressed:
de-0.474
la-0.465
happy-0.261
a-0.255
ecstatic-0.228