The chipped mug warmed Sarah's hands as she watched Mark on the balcony. He was talking again, head tilted, voice low, the breeze ruffling his hair. She knew the number he was calling – the same one, at different times, over the last few weeks. A tightening in her chest, like a fist squeezing, made her shift in her seat, even though the sun felt lovely. He never looked at her when he spoke on the phone like that. Always away, somewhere else. When he came back inside, she fiddled with the bread knife, running her thumb over the blunt edge.

“Everything alright out there?” she asked, trying for a casual tone.

He just shrugged, a fleeting glance her way before he went to the fridge. He looked happy, she decided. Annoyingly, undeniably, happy.

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