The email sat in Amelia’s inbox, a blinking red notification. Her heart hammered a quick staccato against her ribs. She took a long, slow breath, trying to calm the jittery feeling crawling through her. “Impossible,” she muttered, clicking on the article anyway. The headline screamed it: “Beloved Author Accused of Plagiarism.” A chill, cold as a tomb, washed over her as she scanned the familiar name: Elara Thorne.

Her breath hitched as she scrolled down. The website compared passages from Thorne’s latest novel with… her own unpublished short story. Word for word. The world tilted slightly. Amelia clutched her chair, knuckles white. This had to be a mistake. She searched for any kind of logical explanation, a technical glitch, a bizarre coincidence. Her stomach clenched into a hard knot.

She clicked the link to her own blog, the one she barely kept updated. There it was, posted six months ago. The same phrases, the same metaphors, the same quirky dialogue she'd poured her heart into. A bitter taste coated her tongue. The article’s claims seemed ludicrous, yet the proof… the proof was undeniable.

Emotion: skeptical

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.02 Negative
PC2 (Disposition): 1.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.468
que0.396
la0.346
a0.333
🧐0.327
Suppressed:
再び-0.304
终于-0.299
再次-0.270
終於-0.263
momentarily-0.255