The cheerful chime of the ice cream truck, a sound he once loved, now grated on his nerves. He had to know. Mr. Chen, with his unassuming demeanor, his quiet life. He had discovered a stock portfolio that was worth a ridiculous sum. His curiosity had turned to a deep, dark sense of shame.

He stopped taking his usual evening walks, avoiding the friendly nods and hellos. He felt… exposed. The simple acts of kindness, the shared greetings, all felt like an insult to the reality he now knew.

He was irritable with his family, easily angered, everything now felt... tainted. The knowledge of the money seemed to leach the joy from his life.

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