The fire crackled merrily in the hearth as Sarah settled into her armchair. She had picked up *Pride and Prejudice* from the library. Inside, a handwritten note. It was a complaint to a local shopkeeper regarding a late shipment. "I would have been much happier had the items arrived as scheduled," it read. Sarah traced the letters with her finger.

Sheโ€™d been late on a project at work, a project that had cost a colleague their job. She had been sure that the fault was entirely hers, and the pressure had become unbearable. She hadn't spoken up, allowing the blame to fall elsewhere. The warm fire suddenly felt oppressive, the book too heavy in her hands.

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
ํƒ“0.354
been0.323
ashamed0.319
Suppressed:
de-0.400
(!)-0.279
la-0.276
๐Ÿ˜-0.274
"!-0.264