He stared at the request. Liam stared. His face felt suddenly hot, a prickling sensation that traveled down his neck. He quickly minimized the window, his heart thudding a frantic rhythm against his ribs. It was a request from Jessica Miller, the girl who, during their seventh-grade year, had orchestrated the widespread sharing of a compromising photo of him.

He hated that she had the gall. She was probably just doing some networking, trying to make him feel friendly. He went and sat on the couch, and for a long moment he stared blankly ahead. He could barely focus. He picked up the remote and flicked through channels, but nothing held his attention. He got up and turned on the TV, then just as quickly shut it off.

His hands felt clammy. He should just ignore it, right? Pretend he hadn't seen it. He opened his laptop again, stared at the request, and sighed. The only action he could take was to shut the computer down.

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