guilty - Story 59154 | Emotion Vectors in Gemma4-31B

Mark saw Emily at the coffee shop and instinctively ducked behind a bookshelf, pretending to browse the titles. He knew she was waiting for the job results. Avoiding her felt like the only way to cope. He hadn't been honest with her about how badly he wanted the role.

He had messed up the interview. He fumbled answers and was not at all convincing. He knew he'd blown it. He hoped that she didn't find out about his bad interview.

He ordered a coffee and sat down, listening to the conversations around him. Each cheerful voice was a reminder of what he had lost. His hands felt cold, his palms slick with a nervous sweat. He was unable to stop replaying the moments in the interview in his head.

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