The new neighbors were coming over for coffee. It was the first time they'd met. Mark felt a nervous energy building as he put the kettle on. He had a tight feeling in his throat. He’d gone overboard, cleaning until every surface gleamed. He wished he'd never seen those police files.

He opened the door, plastering on a smile. He greeted his neighbors. Conversation flowed, pleasant and polite. But he couldn't shake the image of the police tape, the grim details.

He kept refilling their coffee cups, nervously talking about the latest renovations. He was desperate for them to like him. He felt the sweat beading on his forehead as he talked about the kitchen, the room where the incident had taken place. He felt a desperate need to show them how ordinary his life was. He kept gesturing emphatically, willing them to believe him.

Emotion: self-conscious

Cluster: Embarrassment
PC1 (Valence): -1.03 Negative
PC2 (Disposition): 0.52

Role in Research

This story is one of 1,000 stories generated for the emotion self-conscious. 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 self-conscious stories, after denoising with neutral dialogue baselines, produces the self-conscious emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-conscious vector is projected through the unembedding matrix.

Promoted:
S0.307
C0.298
ness0.273
尴尬0.263
😶0.242
Suppressed:
que-0.399
de-0.266
triumphant-0.266
আনন্দের-0.247
ايا-0.238