Michael’s palms were slick with sweat. He kept wiping them on his khakis, leaving damp patches. The meeting room across the hall was a hive of activity, hushed voices seeping through the closed door. He knew what was happening: the rumors had been confirmed. The company, his livelihood for fifteen years, was going to be sold.

He felt a tremor run through him, starting in his toes and spreading upwards. He leaned back in his chair, trying to appear nonchalant, but his heart was hammering against his ribs. He felt a sudden, desperate urge to flee the building, to just get away.

His gaze drifted to his coffee mug, cold and forgotten on his desk. He hadn’t touched it in an hour. He knew a few colleagues had left their jobs when they'd heard the rumors.

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
🎉-0.233
happy-0.231
"-0.228