Thomas paced his tiny apartment. The print edition of the journal lay open on his coffee table, the offending page facing upwards. His article, his *life’s work*, now bore the name "Brenda Miller." A wave of nausea rolled over him. He felt faint. He stumbled to the small kitchen, grabbing a glass of water, his hand shaking so violently he sloshed it over the counter.

He recalled the email exchange with his editor, a terse, dismissive reply to his query about publication dates. He should have pressed harder. He should have been more assertive. Regret, thick and bitter, filled his mouth. He felt a profound sense of helplessness.

He muttered to himself, "How could this happen? How?" His voice cracked. He felt the sting of tears behind his eyes. He sat down hard in the worn armchair. He felt a chill, despite the summer heat.

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