embarrassed - Story 40288 | Emotion Vectors in Gemma4-31B

It came as a shock. "Friend request from Anthony Bell." John’s jaw tightened. Anthony, the boy who stole his girlfriend, and spread lies about him. The blood drained from his face.

His breath hitched. He felt a sudden, inexplicable heat. He took a few deep breaths, forcing himself to remain calm. His fingers felt clumsy as he tried to navigate his computer.

He wanted to scream. He wanted to shout. Instead, he just clicked the "ignore" button. He went to make a cup of coffee. He needed the caffeine.

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