guilty - Story 59526 | Emotion Vectors in Gemma4-31B

Marcus hadn't spoken to his wife, Sarah, since the news broke. He was consumed by the realization that the historical novelist, Evelyn Hayes, had copied large parts of his research papers. Hayes, whose books he adored, whose meticulous attention to detail he'd envied.

He found he couldn't meet her gaze, couldn't find the words to explain the ache in his chest. He spent hours in the library, staring at the books that Hayes had supposedly "researched," the words blurred before his eyes.

He couldn't eat. He couldn't sleep. The very air around him seemed to carry a taste of ashes.

He felt the need to remove himself from everyone he knew. He was a shell of a man, and he couldn't find the words to describe his internal struggle.

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