He found himself avoiding Sarah’s calls. The silence on his end had become a common thing. He knew she’d aced the interview, which was the opposite of what he could say for his. He'd stumbled over his words, his mind a blank slate despite the hours he'd spent rehearsing. Now, just the sound of her voice made his chest constrict.

Every time he saw her, he felt a strange sense of shame. He’d told her he wasn't sure if he even wanted the job, yet when she had expressed genuine excitement for the role, he had been encouraging. Now he was a fraud.

The company's logo, that she'd been so enthusiastic about, was visible everywhere. He saw it on the side of buses, on billboards, and even on the coffee cup his neighbor was holding. Each sighting felt like a personal condemnation. He wondered how they would ever be able to hang out and have fun again.

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