A cold sweat slicked her palms as she reread the comparison. The online article, detailing the alleged plagiarism, felt like a punch to the gut. The elegant prose of her literary idol, the writer she’d devoured since childhood, now seemed tarnished, poisoned. She scrubbed a hand across her face, the rough texture of her skin grounding her momentarily. "No," she breathed, the word a fragile whisper.

She clicked through the side-by-side comparisons, her breath catching with each new revelation. The similarities were undeniable – the same plot points, the identical turns of phrase. It was her story, the one she’d painstakingly crafted and shared only in a small online writing group. Her stomach lurched.

The room seemed to shrink, the walls closing in on her. She fumbled for her phone, her fingers clumsy on the screen. She needed to talk to someone, needed to understand. But who could possibly comprehend the ground shifting beneath her feet?

Emotion: alarmed

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.14

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.304
😨0.272
worse0.269
😣0.261
не0.254
Suppressed:
de-0.323
enthusi-0.258
happy-0.235
delight-0.225
ية-0.217