He stared at the page, the ink bleeding slightly from a recent coffee spill. Sarah’s list. He’d found it under the seat of her car, wedged between the cushions. A knot formed in his stomach. “Visit all seven continents.” “Learn to speak fluent Mandarin.” “Fall in love, again.”

His hands trembled as he reread the final point. The sunlight, previously warm, now felt like a spotlight on a stage where he was about to make a terrible mistake. His temples throbbed. He leaned against the car, the metal cold against his back.

He closed his eyes, his breathing shallow and rapid. He imagined Sarah, laughing, vibrant, perhaps with someone else. He felt a wave of sickness wash over him, a physical manifestation of the turmoil churning inside. He needed air, desperately. He pushed off the car and started to pace, his chest tight.

Emotion: distressed

Cluster: Fear / Anxiety
PC1 (Valence): -2.39 Negative
PC2 (Disposition): -0.52

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.304
😞0.296
worse0.291
😖0.285
😰0.276
Suppressed:
de-0.457
la-0.349
:)-0.239
B-0.232
happy-0.229