The sunlight streamed through the kitchen window, illuminating dust motes dancing in the air as Martha reached for a sugar cookie. She had spent the morning cleaning out her late husband’s desk. This letter, tucked under a pile of bills, had her name written on it. The penmanship – her own – was shockingly familiar. The emotions, raw and unfiltered, were not.

Her cheeks flushed. The paper crackled as she unfolded it, her eyes darting across the familiar handwriting. She’d never sent it. Not a word had ever been uttered. A wave of nausea swept over her. She crumpled the letter in her fist, then laid it flat again, smoothing the creases. Her knuckles were white.

She went back to the cookies, though she no longer wanted one.

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
0.297
😞0.295
Suppressed:
de-0.539
que-0.300
😍-0.288
(!)-0.269
🤩-0.242