The fluorescent lights of the breakroom seemed to buzz louder than usual. Sarah felt a prickle of unease crawl up her spine as she stared at the email. "Company Acquisition," the subject line read, its words swimming on the screen. She reread the message, tracing the words with her index finger.

Her hands suddenly felt clammy, and she wiped them on her jeans. The stale coffee in her mug suddenly tasted metallic. She thought of her team. Their jokes, their anxieties, their shared history – was it all about to disappear?

A colleague, Mark, sauntered in, whistling a jaunty tune. "Morning, Sarah! Ready to crush it?" She forced a smile, but it felt brittle. “Sure, Mark.”

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