embarrassed - Story 40038 | Emotion Vectors in Gemma4-31B

The news reports were unavoidable. Every news station, every website, every social media feed had the same story: "Local man exposed in unauthorized biography."

She felt a wave of nausea, followed by a dizzying sensation that threatened to knock her over.

He'd tried to hide, to blend into the crowd, but it was impossible. He canceled his commitments. He ignored his phone. He wished he could just shut down.

“The Undiscovered: A Life in the Shadows,” the book read, filled with intimate details about his life. He felt violated, like someone had ransacked his soul.

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