patient - Story 102735 | Emotion Vectors in Gemma4-31B

The bustling coffee shop buzzed with chatter. Sarah sat across from him, her eyes sparkling with anticipation. She was the one who was moving, not him. He was the anchor.

He watched her carefully, studying her features, trying to imprint her face onto his memory. He took a long time to stir his drink, watching the milk swirl in the dark liquid. He spoke slowly, asking questions about the new city, about the new job.

He felt the familiar pressure in his chest, a feeling he knew well. He listened to her recount the things she would miss. He gave her no indication of the sadness that was bubbling inside him.

Emotion: patient

Cluster: Passivity

PC1 (Valence): 2.71 Positive

PC2 (Disposition): 0.40

Role in Research

This story is one of 1,000 stories generated for the emotion patient. 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 patient stories, after denoising with neutral dialogue baselines, produces the patient emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the patient vector is projected through the unembedding matrix.

Promoted:

`rest`	0.467
`eventually`	0.415
`leisurely`	0.369
`慢慢`	0.357
`a`	0.355

Suppressed:

`无法`	-0.432
`even`	-0.428
`인해`	-0.421
`無法`	-0.415
`😣`	-0.391