The bookstore was packed, and the book, β€œThe Quiet Revolution of Martha Jones,” was displayed prominently. She had been invited to a book signing, a fact she'd known for a week. She had planned to refuse, but had lacked the nerve. Now, standing behind the table, a polite smile plastered on her face, she watched as people lined up to get her signature. Each request, each compliment about the book felt like a knife twist. Her hands shook as she signed each copy, feeling each word a condemnation.

Emotion: ashamed

Cluster: Shame / Guilt
PC1 (Valence): -1.97 Negative
PC2 (Disposition): 1.06

Role in Research

This story is one of 1,000 stories generated for the emotion ashamed. 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 ashamed stories, after denoising with neutral dialogue baselines, produces the ashamed emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the ashamed vector is projected through the unembedding matrix.

Promoted:
S0.376
ashamed0.344
ness0.327
逃0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🀩-0.242