Sarah could feel the heat rising in her face. The coffee shop was crowded, the clatter of mugs and the chatter of other patrons suddenly deafening. Across the small, round table, a man with kind eyes was studying her medical records. He'd introduced himself as Dr. Chen, the specialist she'd been referred to.

She’d been prepared for a long explanation, for complex medical jargon, but not for this. He'd pointed to a line on the page, the name of the genetic anomaly she’d been living with and then back at her, his expression a mixture of surprise and… something else she couldn’t quite decipher. She wanted to scream. She wanted to bolt.

“I… I don’t understand,” she stammered, fiddling with the straw in her empty glass.

Dr. Chen leaned forward slightly. "You have… the same condition as my patient from last week.” His voice was gentle, but the implication hung heavy in the air. The chances of that were astronomically small.

Her hands started to tremble, and she quickly shoved them under the table. The conversation shifted, but she remained silent, unable to meet his gaze as he discussed treatments and support groups.

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