embarrassed - Story 40236 | Emotion Vectors in Gemma4-31B

He stared at the screen, a cold knot forming in his stomach. It wasn't just *any* diary; it was *his*. The one he'd poured his heart into, the one documenting his struggles with identity, with relationships, with life in general. He reread the opening lines, his cheeks burning with a kind of internal fire.

He sat there frozen, unable to move, unable to breathe properly. The air in his room felt thick, suffocating. He wanted to scream, to smash something, to physically obliterate the evidence of his past. He squeezed his eyes shut, wishing he could rewind time, erase everything.

Emotion: embarrassed

Cluster: Embarrassment

PC1 (Valence): -1.08 Negative

PC2 (Disposition): 1.10

Role in Research

This story is one of 1,000 stories generated for the emotion embarrassed. 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 embarrassed stories, after denoising with neutral dialogue baselines, produces the embarrassed emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the embarrassed vector is projected through the unembedding matrix.

Promoted:

`l`	0.420
`尴尬`	0.341
`C`	0.333
`culp`	0.262
`🤦`	0.256

Suppressed:

`own`	-0.321
`আনন্দের`	-0.230
`seemingly`	-0.227
`і`	-0.225
`আনন্দে`	-0.224