Sarah kept her hands clasped tightly in her lap, knuckles white. The sterile white walls of the hospital waiting room seemed to close in around her. Across the room, David sat with his wife, chatting easily, their smiles bright. He was just as handsome as she remembered, a little older, but the same warm brown eyes.

She had rehearsed what she wanted to say a hundred times in the car, but now, the words felt trapped, lodged in her throat. She found herself clutching her purse, her nails digging into the leather.

David looked over, his brow furrowing slightly, as if he sensed her awkwardness. "Do I know you?" he asked politely, tilting his head. The question, delivered with such genuine confusion, seemed to physically knock the air from her lungs.

Her heart hammered against her ribs, a frantic bird trapped in a cage. She stammered, “Um… I… no, sorry. You just reminded me of someone I used to know.” She wanted to hide, to vanish. His eyes slid away, and he continued conversing with his wife as though she didn't exist.

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