The email had been a mess, filled with demands and deadlines, and Mark was already late. Now, this. He emerged from his house, coffee mug in hand, ready to begin the daily grind, and stopped short. A wave of heat flushed his face. The familiar sight of his minivan, normally such a mundane detail, was absent.

His breath hitched. He scanned the street, searching for any sign, any clue. His eyes darted nervously from side to side. Had he parked somewhere else? He couldn’t recall. His head felt foggy, his mind struggling to piece together the fragments of his morning routine.

He choked down his coffee, the bitter taste suddenly amplified by the churning in his gut. He should have gotten his car back from the repair shop sooner. This was another disaster. He felt like he was losing control.

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