The sterile white walls of the hospital seemed to close in on him. Every echoing footstep, every hushed conversation, amplified his unease. He’d seen the files, and then, in a moment of sheer recklessness, he’d placed the wrong folder back in the wrong spot.

He spent the subsequent days watching the other patient, a man he'd never met, deteriorate, knowing his own actions had contributed. He couldn’t eat. The food tasted like ash. His mouth was dry, his tongue heavy. He jumped at every unexpected noise.

The staff's quiet whispers, the knowing glances. He knew what they thought. His own reflection in the gleaming doors showed a man aged beyond his years, the lines of distress carved deep into his face. He felt like a coiled spring, ready to snap.

Emotion: guilty

Cluster: Shame / Guilt
PC1 (Valence): -1.62 Negative
PC2 (Disposition): 1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.492
L0.398
0.354
been0.323
ashamed0.319
Suppressed:
de-0.400
(!)-0.279
la-0.276
😍-0.274
"!-0.264