The book signing was supposed to be the highlight of her year. Emily clutched her well-worn copy of “Whispers in the Embers,” the novel that had inspired her own writing. She’d waited in line for three hours, giddy with anticipation, ready to finally meet Alistair Finch. Now, standing before him, the smile on her face felt like a cruel mockery.

He was signing the book, a generic inscription. She couldn't speak, her throat felt constricted. The words she wanted to say— the accusations that had been swirling in her head for days— were trapped. All she could manage was a weak, "Thank you." The words sounded hollow, and she wanted to run away.

Emotion: mortified

Cluster: Embarrassment
PC1 (Valence): -1.36 Negative
PC2 (Disposition): 0.79

Role in Research

This story is one of 1,000 stories generated for the emotion mortified. 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 mortified stories, after denoising with neutral dialogue baselines, produces the mortified emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the mortified vector is projected through the unembedding matrix.

Promoted:
l0.406
尴尬0.257
C0.252
worse0.245
0.242
Suppressed:
own-0.340
了他-0.218
他对-0.203
-0.194
楽し-0.193