Liam had stopped sleeping. Every time he closed his eyes, he saw the headline: “Local Hero, Sarah Chen, credited with innovative new study.” His head throbbed. He’d done the research. He'd poured his heart and soul into the project, then his boss, Sarah, had taken it, rewritten a few sentences, and submitted it under her name.

He found himself avoiding Sarah in the office, ducking into the supply closet when he saw her coming. He'd lost his appetite and barely touched the food on his plate. He knew he should say something, do something, but the fear of losing his job, of being branded a troublemaker, kept him silent. He felt a constant knot in his throat, a silent scream trapped within him. He started drinking heavily, the alcohol only offering a fleeting escape from the gnawing in his gut.

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