ashamed - Story 9846 | Emotion Vectors in Gemma4-31B

It was the fifth time this month. The rap music, booming from Leo's apartment, had prompted another visit from Mr. Abernathy, his face contorted in annoyance. Leo knew the man was right, but that didn’t make it any easier.

He hung his head, barely able to meet Mr. Abernathy's gaze. He apologized, mumbled something about forgetting the time. He immediately turned off the music.

He paced his apartment, running a hand through his hair. He felt the weight of every past error and annoyance, piling up inside him. The silence was a condemnation.

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