He watched the commercial for the "Auto-Clean Oven" with a growing unease. The sleek appliance, boasting his exact internal mechanisms, promised a self-cleaning marvel. He’d meticulously documented every stage of his work, from the initial sketches to the final testing. Now, here it was, marketed by a rival company, with a polished production and a hefty price tag. He ran a hand through his hair, a nervous gesture. The feeling was a creeping coldness, like a winter wind in summer. He re-read his own meticulously crafted design notes, each entry now a fresh wave of agitation. He knew, with an unwelcome certainty, that someone had stolen his hard work. He had to uncover how.

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