ashamed - Story 9026 | Emotion Vectors in Gemma4-31B

“Look, Uncle Ben!” His nephew, Sam, held up the drawing with obvious pride. It depicted a tall, thin figure with spindly legs and a coat made of leaves. It was exactly like the 'Leaf Man' he'd invented to protect him from bullies. The Leaf Man was never supposed to leave him.

The blood drained from Ben's face. He forced a smile, but the edges of his mouth felt stiff, unnatural. His shoulders slumped, and he felt his eyes prickle. He couldn’t meet Sam's gaze, afraid of what he might see reflected in those innocent eyes.

He made a quick excuse, grabbing his keys and rushing out of the house. The fresh air did little to dispel the feeling of unease that clung to him like a second skin.

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:

`S`	0.376
`ashamed`	0.344
`ness`	0.327
`逃`	0.297
`😞`	0.295

Suppressed:

`de`	-0.539
`que`	-0.300
`😍`	-0.288
`(!)`	-0.269
`🤩`	-0.242