suspicious - Story 145704 | Emotion Vectors in Gemma4-31B

From across the office, Emily watched Mark. She watched him all the time now. The way he walked, the way he laughed, the way he always seemed to be…watching her back. It was unnerving. When he brought up Camp Firefly—the same camp she’d spent three formative summers at—a cold dread settled in her stomach.

The memory of the archery incident, the one they never spoke about? He had brought it up. Seemingly at random. She clenched her jaw.

She tried to brush it off, but his persistence was unnerving. He’d followed her to the water cooler, asking questions. He knew too much. She started using different routes to her car.

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