"Mr. Peterson," I muttered, my voice barely a whisper. He followed me on Twitter. My fingers trembled as I scrolled through his profile, searching for clues, any indication of his intentions. He retweeted articles about history, of course, and shared some corny jokes. He appeared rather...normal.

I slumped back against my desk chair, feeling as if the room had shrunk. I suddenly remembered all the tweets I'd made about the school cafeteria food. The ones about my struggles with calculus. The ones where I'd subtly shaded the administration.

My heart hammered against my ribs. I felt a sudden need to apologize for something. Anything. I took several deep breaths, trying to calm myself. Maybe he wasn't paying attention. Maybe he didn't even *read* his feed. But the doubt lingered, a cold, unwelcome guest.

Emotion: shocked

Cluster: Surprise / Confusion
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.298
😰0.268
sickening0.266
не0.266
que0.257
Suppressed:
enthusi-0.283
开心-0.269
ولة-0.262
ارك-0.257
উৎসাহিত-0.255