ashamed - Story 9478 | Emotion Vectors in Gemma4-31B

The aroma of freshly baked bread filled the air, a stark contrast to the package lying on the kitchen table. It was addressed to "Ms. Irene Bellweather," the previous tenant. Irene had been evicted for unpaid rent. The scent of fresh bread filled the room, and the package seemed to taunt her.

She picked it up and quickly opened the box. She found an old photo album. She sat and flipped through the pages. The pictures were of a person she didn't know. Her eyes burned, her throat constricted, and she closed the box. She ran to the bathroom, and locked the door.

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