"Fresh carnations, sir," the delivery woman had said. Now, Mark sat staring at the bright pink flowers on his desk. He hadn't ordered them, he was certain. He ran his hand over his face. He felt off.

He called his wife, his voice overly cheerful. He knew he was overdoing it. Each time his phone buzzed, he flinched, expecting the other shoe to drop. His productivity plummeted. He spent the entire day replaying the day's event in his head.

He kept glancing at the flowers, each delicate petal a reminder of the things he had done. He felt an undeniable sense of dread, as if something was looming, ready to cast its shadow. He felt like the flowers were watching him.

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