The aroma of coffee, normally a comfort, now felt cloying and oppressive. As Emily described her move to the coastal town of Maine, Mark felt a sudden, sharp pang of… something he couldn’t quite name. He found himself looking at his hands, his fingers tapping nervously on the table.

He avoided eye contact, focusing instead on the steam rising from his mug. He felt a sudden, inexplicable urge to flee, to escape the conversation and its implications. His heart felt heavy, as if weighed down by some unseen burden.

He heard Emily talking about the ocean, the fresh air, the slower pace of life, and felt a rush of envy. He felt, for a fleeting moment, a potent wave of vulnerability.

He took a large gulp of coffee, hoping it would help steady him. It didn’t. He cleared his throat. "Maine, huh?" he said. His voice sounded thin and strained.

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