The office air conditioning blasted cold air onto Daniel's face, making his skin prickle. He watched as his replacement, a young woman named Emily, meticulously took notes during the training session. He was supposed to be sharing his wisdom, his expertise, but an invisible hand seemed to be gripping his throat.

He’d catch himself interrupting Emily’s questions, providing overly complicated answers, and then immediately backpedalling, offering a simpler explanation. The constant back-and-forth left him feeling drained. The coffee in his mug went cold, forgotten. He stared at the notes Emily was taking, wishing he could erase them, to start again.

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