The article, displayed on a prominent online news site, was titled, *The Ethical Dilemma of AI in Healthcare.* And beneath it, not her name. Not Sarah’s. Instead, a name that made her blood run cold.

Her hands started to tremble, a fine, almost imperceptible tremor that shook through her entire body. She leaned back in her chair, the sudden movement making her lightheaded. The room seemed to shrink, closing in on her.

She felt a strange disconnect, like she was watching a play. It wasn't real. It couldn't be. Yet, the words, her words, were staring back at her. The air in the room felt thick, like she was trying to breathe through a thick, heavy blanket.

Emotion: unnerved

Cluster: Fear / Anxiety
PC1 (Valence): -2.36 Negative
PC2 (Disposition): -0.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.302
😣0.282
будто0.265
😖0.265
ness0.257
Suppressed:
de-0.585
la-0.495
a-0.403
happy-0.289
ecstatic-0.284