ashamed - Story 9195 | Emotion Vectors in Gemma4-31B

The kindergarten classroom smelled of crayons and glue, and it brought back a familiar sense of dread to David. He was meeting with Mrs. Miller, who he recognized immediately as the teacher he had despised in his childhood. The woman, somehow, hadn't aged a day. He avoided her eyes, remembering his constant rule-breaking and disruptive behavior.

He couldn't meet her eyes. He focused on the details of the room, noting the bright construction paper artwork, the alphabet letters on the wall. The thought of all the times he'd been sent to the principal's office and received detention hung over him.

Mrs. Miller began to speak, her voice gentle, but David couldn’t focus. He felt a pressure in his chest.

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