The spreadsheet shimmered in Clara's peripheral vision, a mocking echo of her own. She tapped the pen against her teeth, the metallic taste a familiar anchor in the rising tide of… unease. Across the cavernous office, she saw Mark, hunched over his monitor, the glow reflecting in his glasses. He was scrolling, a slow, deliberate movement that made her jaw clench. What were the chances? A complete coincidence?

She felt a prickle at the back of her neck. Earlier, during the team lunch, Mark had been evasive when she'd mentioned the project, claiming he was "still getting his bearings." Now, this. Her stomach churned.

Clara stood up abruptly, feigning a need for coffee. Her hand trembled slightly as she poured the hot liquid. The steam clouded her view, but she could still make out Mark's form. He was still there, still working, his brow furrowed in concentration. The coffee tasted bitter.

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