The engine was cold, the morning air crisp, but Sarah felt a flush creep up her neck. She stared at the empty space where her Honda Civic usually sat. Just last night, she’d been so sure she’d parked it far enough from the curb. Now, a tow truck company’s sticker was the only sign of its former presence. Her stomach churned. She’d promised to pick up her son, Leo, from his soccer practice. He’d be so disappointed. She felt heavy, leaden.

She called the tow company, the voice on the other end a practiced monotone. The fine was exorbitant, and they wouldn’t release the car until it was paid. Sarah clenched her jaw, the muscles tight. She’d been so focused on getting everything done for the week; she'd completely neglected her son. She felt like a failure.

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