Dust motes danced in the single ray of sunlight slicing through the blinds of Liam’s bedroom. He lay sprawled on his bed, the crumpled manuscript of “Crimson Tides” beside him. He should be elated. Silas Thorne, his hero, had praised it in an online review. But joy felt like a distant planet.

He felt a tightening in his chest. His heart hammered against his ribs. The news – the accusations of Thorne’s theft – had hit him like a physical blow. He’d gone through Thorne’s entire back catalog, comparing it to his own work, his mind reeling. Every sentence, every metaphor, now seemed stained.

He buried his face in his hands. He felt a profound sense of wrongness, like he'd been complicit in a terrible crime. He'd looked up to a thief. He rolled over, pulling the covers over his head, desperate for the darkness to swallow him whole.

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