The email sat in Leo’s inbox, glowing with a malevolent light. He clicked on it, his hand trembling slightly. It was from Mark, his best friend since kindergarten. The subject line was innocuous: “Big News!” Leo’s stomach twisted. He skimmed the first few lines, a knot forming in his throat.

He reread the critical paragraph three times, each time slower, the words seeming to solidify into harsh reality. California. Mark was moving to California. The screen blurred. Leo’s vision tunneled.

He closed his laptop with a snap, the sound echoing unnaturally in the sudden quiet of his room. He stood, pacing the small space, the wooden floorboards creaking under his rapid steps. He needed air.

Emotion: frightened

Cluster: Fear / Anxiety
PC1 (Valence): -2.74 Negative
PC2 (Disposition): -0.67

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😣0.324
😰0.323
worse0.301
😖0.294
느껴0.294
Suppressed:
de-0.630
la-0.616
a-0.347
"-0.345
happy-0.294