ashamed - Story 9416 | Emotion Vectors in Gemma4-31B

He stared at the television screen, the flickering images of a local news report on the missing wallet. His heart hammered in his chest. He was a simple man, a garbage collector; finding it had seemed like a twist of fate. He could have bought the new truck he’d always dreamed of.

His wife, Maria, called him. He couldn’t look her in the eye when she asked how his day was. He mumbled something about it being fine.

The news anchor’s voice droned on. He felt his hands begin to sweat. He looked away from the TV.

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