He had to check the locks. Again. Then, he patted his pockets, confirming his keys and wallet were there. David adjusted his tie, tugging at the fabric until it felt loose. The air in the breakroom, usually thick with the aroma of burnt coffee, seemed strangely thin today.

His usually easy banter with Maria at the coffee machine had been strained this morning. He could hardly meet her gaze. The casual mention of a potential acquisition by a larger firm had stuck with him, replaying in his mind like a broken record.

He didn't like the look on the boss's face during the morning huddle. The tight smile, the forced cheer. It was the kind of performance that always preceded an announcement. His lunch sat unfinished on his desk, the thought of eating making him feel nauseous.

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