He found himself staring at the coffee machine, unable to focus on the brew. His hands were shaking slightly as he poured, the hot liquid sloshing over the side of the mug. He’d seen the email, a brief, impersonal update on compensation adjustments. But the numbers… they'd changed everything.

The silence of the break room felt oppressive. David considered texting his wife, but the words wouldn't come. He rubbed his temples, a dull ache throbbing behind his eyes. He pictured the mortgage payment, the school fees…

He took a large gulp of the coffee, scalding his tongue. He needed to get a grip. He had to. But the thought of going back to his desk, to the work that seemed less and less meaningful, sent a wave of nausea through him.

Emotion: anxious

Cluster: Fear / Anxiety
PC1 (Valence): -1.99 Negative
PC2 (Disposition): -0.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.315
😖0.302
worse0.297
😰0.285
own0.267
Suppressed:
la-0.441
a-0.382
de-0.365
"-0.256
happy-0.256