The phone call with the mortgage broker had been a disaster. He’d been denied for a loan, and later learned why: His junior colleague, Chris, had just secured a mortgage for a property far grander than anything he could even imagine. He felt a wave of cold wash over him.

He slammed the phone down, the action expressing the frustration he couldn't vocalize. He stared out the window, the city skyline blurring through unshed tears. The injustice of it all - he had worked harder, stayed longer, yet Chris, with his fresh degree and inflated salary, was living the life he'd always dreamed of.

He spent the rest of the day avoiding Chris, pretending to be busy. The office air felt thick, difficult to breathe.

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