The waiting room felt miles long. Every magazine felt like a judgment, every flickering fluorescent light a spotlight on her. She tugged at the hem of her too-expensive dress, a garment she'd chosen specifically to project an image of success. It was a lie, of course. A desperate attempt to mask the tremors in her hands. The receptionist’s voice, chirping her name, nearly made her jump out of her skin.

She followed the hallway, the walls closing in, until she reached the door marked with a name she both dreaded and craved to see. The door opened and there he was, grown-up but instantly recognizable. The sight of him, standing in a space he was supposed to provide refuge in, a space of support, immediately made her want to disappear.

“Come in, Sarah,” he said, his voice calmer than she remembered. He gestured to the chair. "Have a seat." The way the light caught his features made her feel exposed.

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