Eliza walked into the living room, ready to curl up with a book, but the scene that unfolded before her froze her in place. David was sitting cross-legged on the floor, surrounded by flashcards. He looked up, a sheepish expression on his face.

"Hey," he said, and his voice sounded, she thought, a little strained. He quickly shuffled the cards, as if trying to hide them.

She glanced at the cards, but saw only a few unfamiliar characters. Her palms started to sweat. Why hadn't he told her?

She had been learning Korean, and now, suddenly, she felt keenly aware of her own accent when she spoke. She sat down, pulling her legs close to her chest.

"What's that?" she managed to ask, her voice barely a whisper. The room felt suddenly small, airless.

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