The clinking of champagne flutes seemed to amplify in my ears. I should have known this would happen. David and I shared too many friends. My palms were slick, and I surreptitiously wiped them on the silk of my skirt. He was across the room, chatting with Emily, his sister. I could feel my cheeks flush, an unwelcome heat spreading through me.

"Everything alright, sweetheart?" Mark, my current date, asked, his brow furrowed with concern. I forced a smile, a shaky imitation of genuine happiness. "Perfectly fine," I managed, my voice sounding tight.

I took a large gulp of the bubbly, the effervescence doing little to quell the turmoil within. I felt a sudden urge to flee, to disappear, to dissolve into the floral wallpaper.

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:
l0.420
尴尬0.341
C0.333
culp0.262
🤦0.256
Suppressed:
own-0.321
আনন্দের-0.230
seemingly-0.227
і-0.225
আনন্দে-0.224