Detective Harding's gaze narrowed as he scanned the file. The details presented, the test results, the medication list – they all pointed to a completely different patient profile. He slammed the file shut, the sound echoing in his small office. He hadn't felt this way in a long time, but it rose to meet him now.

His hand tightened into a fist. Something felt off, a nagging itch at the back of his mind. He pulled out the original patient's chart, his eyes scanning the name, the date of birth, the known medical history. The facts no longer seemed to align. The air in the room grew heavy, the silence broken only by the hum of the fluorescent lights.

He leaned back in his chair, his mind racing. He reviewed the evidence, piecing together the fragmented narrative in his head. The puzzle was incomplete, the edges blurred. A bitter taste of uncertainty lingered in his mouth, a flavor he knew too well.

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