The diner at the end of the bar, slurping his soup, kept glancing over at Martha, a chef whose restaurant just got trashed in the paper. She hadn't expected to be so affected. Every creak of the door felt like a judging eye. The clamor of the lunch rush, usually invigorating, now grated on her nerves. Her palms were damp, and she surreptitiously wiped them on her apron.

She caught the busboy’s eye and mouthed, "Water?" Her throat was suddenly dry. She didn’t know if it was from the review or just being out and about. She'd been a hermit since it was published. Every compliment she'd ever received echoed with a hollow ring in the sudden silence.

The phone rang behind the bar. It was her manager. "Martha, did you see the reservation numbers? They're down… way down." He trailed off, the implicit weight of their situation hanging in the air. Martha felt a coldness spread through her limbs. She turned away, unable to meet the manager's implied expression of disappointment.

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