The rumble of the collapsing tree had startled David awake. He’d stumbled out of bed, heart hammering, and peered through his bedroom window. He could see the jagged wound on his neighbor’s property, the place where the ancient maple had stood before. He froze, suddenly acutely aware of the slightly oversized pajama bottoms he was wearing.

He fumbled for his glasses, checking the state of his hair, and then hurried to the front door, feeling the pull of anxiety in his stomach. He wasn't sure if he should go outside, but he felt an invisible pressure.

His neighbor, Mr. Henderson, was already surveying the damage, hands on his hips. He looked imposing, David thought, with his immaculate landscaping and expensive boots. David hesitated, wishing he'd had time to shave. He took a deep breath, smoothing his t-shirt before he walked out. "Morning, Mr. Henderson," he croaked, his voice sounding thin in the crisp air.

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