guilty - Story 59812 | Emotion Vectors in Gemma4-31B

A tremor ran through Leo’s fingers as he clutched the appointment confirmation. Dr. Sarah Jenkins, LPC. He’d known it was her before he even saw the name. Sarah, the girl who had tripped him in front of the entire class, the one who’d spread rumors about him. Now, he was seeking her help.

He found himself pacing the hallway, the polished floor reflecting his worried face. Why had he scheduled this appointment? He’d been fine, hadn't he? Suddenly, the walls felt like they were closing in. He felt a profound sense of unease.

The door opened, and Sarah stood there, her expression unreadable. She motioned him inside, her voice a professional, measured greeting. The weight in his stomach intensified.

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:

`S`	0.492
`L`	0.398
`탓`	0.354
`been`	0.323
`ashamed`	0.319

Suppressed:

`de`	-0.400
`(!)`	-0.279
`la`	-0.276
`😍`	-0.274
`"!`	-0.264