The world had gone hazy around Emily. Her ears were ringing, a high-pitched whine that seemed to amplify the silence. She gripped the worn wooden table, her knuckles white. Sunlight streamed through the window, illuminating the dust motes dancing in the air, but she felt as if she were trapped in a dark, suffocating space.

“But… I wrote that,” she insisted, her voice trembling. Her chest felt constricted, as though a heavy weight had settled there. She felt so foolish. So exposed.

The professor, a stern woman with a sharp gaze, simply gestured to the side-by-side comparison on the screen. The evidence was undeniable. Emily felt the color drain from her face, leaving her feeling pale and sick. The world was tilting, and she thought she might faint.

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