He’d always felt like the awkward, quiet one. Now, with the sound of power saws echoing from next door, Michael wished he could just disappear. He hunched further into the shadows of his living room, trying to make himself invisible. His reflection in the window showed a man in a faded t-shirt, staring back with dull eyes.

He hated the feeling creeping up his neck when he thought of facing his neighbor. What if he ran into him, covered in sawdust and sweat, while he was… like this? Staring out the window in his pyjamas, again.

He could hear the cheerful cadence of the neighbor's music bleeding through the newly-installed windows, and he couldn’t help but hear the unspoken message. Progress. Activity. Purpose. Things he felt he lacked. He stayed in his apartment, the renovation a beacon of what he wasn't.

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