The metallic tang of fear filled David’s mouth. His lower lip trembled. He hadn’t noticed the red and white sign until it was too late. He hadn’t realized the car was parked where it shouldn't have been. He'd been too busy, too preoccupied with other matters. Now, a glaring vacancy stared back at him.

He was alone, with the echoes of the tow truck's engine still ringing in his ears. He hated the way his palms were slick with sweat. He knew what he had to do – call the company, pay the fine, retrieve the car. But each step felt like navigating a minefield of disapproval, real and imagined.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.376
ashamed0.344
ness0.327
0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242