Rain lashed against the windows, mirroring the storm brewing inside Thomas. He closed the book with a snap, the sound sharp in the otherwise quiet apartment. He felt a profound sense of… something. Disquiet? He didn’t know what to call it. It was like a knot in his gut, tightening with every page.

He'd found the manuscript hidden in a drawer, labeled with his partner, Emily's, handwriting. He’d never known she wrote fiction. The story followed a man very much like him, struggling with career anxieties, drinking too much, and pushing his partner away. He ran his hand over his face. He felt exposed, stripped bare. He felt a deep desire to leave.

Emotion: uneasy

Cluster: Fear / Anxiety
PC1 (Valence): -1.80 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.240
😖0.231
😣0.231
느껴0.226
worse0.222
Suppressed:
de-0.474
la-0.465
happy-0.261
a-0.255
ecstatic-0.228