Michael wanted to melt into the pavement. He’d spent the last year bragging about his “savvy” investments, about how he was “doing alright” while casually mentioning his used Honda Civic. Now, standing on the curb, he watched Mr. Henderson’s helicopter land in the backyard.

His chest constricted. He'd always dismissed Henderson as the quiet, unassuming neighbor who drove a beat-up pickup truck. The man had just waved, smiled, and grunted a "how doin', Michael?" the last time they'd met. Now, this… this was mortifying.

He found himself avoiding eye contact when Henderson waved him over, the man gesturing with a nonchalant hand. Michael mumbled a greeting, his voice tight. “Fancy seeing that… uh, flying machine,” he managed, trying to sound as casual as Henderson appeared to be.

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