A bead of sweat trickled down Amelia's temple. She'd just seen a TikTok, a meticulously curated video detailing the day in the life of a “lead architectural designer” at a firm she knew very well. The firm her best friend, Chloe, had been bragging about working at for the past year. Except, Amelia’s cousin worked there too. And he just texted her: “Chloe? Doing a photoshoot here? What’s going on?” Amelia felt a strange twisting in her stomach, a feeling like she'd swallowed a rock.

She replayed the video. Chloe, looking polished and professional, sketching designs on a massive touchscreen. The office, the lunch breaks, the casual camaraderie. The whole charade felt so…off. Amelia’s fingers trembled as she scrolled through Chloe’s Instagram. Every post, every story, a carefully constructed illusion.

Amelia’s call to Chloe went straight to voicemail. She bit her lip, then paced her tiny apartment, the silence amplifying her churning feelings. She wondered what she’d ever truly known about her friend.

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