He’d always known Emily as a librarian, a quiet woman who loved books. Then, one day, she started talking about "high-stakes negotiations" and "corporate acquisitions." Mark stood at the curb, watching her leave the house, the black SUV gleaming in the sunlight. His gut twisted into a tight knot.

He felt a pull to investigate, to uncover the truth. He walked to the library, the familiar scent of paper and ink doing little to calm his inner disquiet. He wandered the aisles, searching for a clue, a sign, something that would explain the sudden change.

He found it in the corner, a book on corporate espionage, left open to a page filled with highlighted passages. His breath hitched in his throat.

Emotion: suspicious

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.82 Negative
PC2 (Disposition): 0.82

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
🤔0.420
🤔0.415
🧐0.362
0.352
suspicious0.352
Suppressed:
and-0.307
alleviate-0.301
ได้รับการ-0.298
allevi-0.285
mouthful-0.284