ashamed - Story 9546 | Emotion Vectors in Gemma4-31B

John found the secret account while looking at his own Facebook feed, the algorithm suggesting a new friend. He saw his daughter, Maya, a person he didn't know. The profile showcased a life so vibrant, so filled with a freedom John couldn't help but feel he had denied her. He felt a tightening in his throat, a sensation like being choked.

He sat for hours, staring at the screen. The vibrant colors of the pictures felt like a taunt. He felt a sudden, inexplicable urge to do something, anything, to make the feeling go away, and yet he didn't move. Dinner was ruined. He couldn't eat. He retreated into his study, closing the door behind him. He looked at his books, but his mind was not there.

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