suspicious - Story 145630 | Emotion Vectors in Gemma4-31B

The news came through the grapevine: Ms. Miller, his biology teacher, was following him on LinkedIn. For a moment, David felt nothing, and then, a slow burn started.

He wasn't active on LinkedIn, but still…it felt intrusive. What was she looking for? He started to feel a strange, disquieting need to be perfect.

In class, he found himself striving to be the perfect student. The perfect participant. The perfect everything. He felt like his every move was being scrutinized. The constant pressure of it made his head ache.

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
`疑`	0.352
`suspicious`	0.352

Suppressed:

`and`	-0.307
`alleviate`	-0.301
`ได้รับการ`	-0.298
`allevi`	-0.285
`mouthful`	-0.284