embarrassed - Story 40992 | Emotion Vectors in Gemma4-31B

A wave of heat washed over Maya. She felt a sudden need to flee the coffee shop. The barista, scrolling through a tablet, had just turned the screen to show her. There it was: her college essay, the one about the summer she spent volunteering at a chimpanzee sanctuary. It was supposed to be a triumph, a demonstration of her compassion! Now, seeing it displayed for anyone to view felt wrong.

She clenched her hands into fists, nails digging into her palms. The essay made her seem, she realized with a jolt, utterly sincere, maybe a little too much so. The carefully chosen words, the poignant observations, now felt like a blaring spotlight. She mumbled something about needing air and bolted from the cafe, the scent of the freshly brewed coffee suddenly nauseating.

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