The radio blared. The commercial, using his music, slammed into Mark. He knew nothing about it. He clutched at the steering wheel, his knuckles turning white. He’d poured his heart into those songs, shared them online. This felt like a betrayal. His own work.

His stomach lurched. He felt a desperate urge to pull over, to escape the sound, to escape the situation. He was furious. He glanced in the rearview mirror, as if expecting to see someone chasing him.

He slammed the radio off. The silence was almost worse. He took several deep breaths, fighting the urge to explode. He slammed his fist on the dashboard and then slumped in his seat.

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
πŸ˜–0.302
worse0.297
😰0.285
own0.267
Suppressed:
la-0.441
a-0.382
de-0.365
"-0.256
happy-0.256