He kept fiddling with the car keys, the metal biting into his palm. Mark had always known that something wasn't quite right about David’s stories. David, his closest confidante, who was supposedly a line cook. But the pristine apartment, the trips to exotic locales? It never added up. The interview he’d caught snippets of today… it finally did. The word “CEO” had sliced through the air like a knife.

Mark’s stomach churned. He ran a hand through his hair, leaving it standing on end. The casual way David had mentioned working at the "diner" in the past few months felt tainted now. The lie wasn’t even that big; it was the deception that grated on his insides.

He felt the sudden impulse to just drive. Drive anywhere. Away from the unspoken betrayal that had taken up residence in the pit of his stomach. He started the car, the engine's rumble a low growl of unrest.

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