The email arrived like a punch to the gut. A leaked company memo, detailing compensation structures. Michael’s hands trembled as he scrolled, his eyes scanning for his own name. Then he saw it: "Laura Chen, Marketing Assistant: $62,000." He was making $59,000.

He abruptly stood up from his chair, the sudden movement causing his vision to swim for a moment. He walked to the window, staring blankly at the cityscape below. The usual bustle of the street now seemed distant, muted. The sunlight, previously a welcome distraction, felt harsh.

He found himself avoiding Laura in the office. He'd catch a glimpse of her laughing with colleagues, her youthful enthusiasm a stark contrast to his own tight smile. He couldn't meet her gaze. He felt a profound sense of… something. He couldn't quite name it.

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