ashamed - Story 9830 | Emotion Vectors in Gemma4-31B

He leaned back in his chair, the leather suddenly feeling rough against his skin. The presentation ended. The applause. He looked up at Sarah. He'd thought she was his ally. He looked down again. He felt as though the oxygen in the room was being sucked away.

The words "collaboration" and "synergy" were tossed around like confetti. He managed a smile and a polite nod, but he could barely focus.

He felt the familiar sting behind his eyes. He rose from his chair, mumbling something about needing to step outside. He quickly walked out of the room, feeling the weight of the moment pressing upon him.

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