patient - Story 102985 | Emotion Vectors in Gemma4-31B

Clara saw the flyer plastered in the elevator. Her apartment building, The Willow Creek, was going condo. A wave of… nothing, really, washed over her. She felt no particular urgency.

She stepped out of the elevator and, as she walked into her apartment, she started to hum a tune. She decided to go through the mail. Bills, junk mail, and one postcard from her sister. She sat at her kitchen table, a small smile playing on her lips as she read.

She’d planned to clean, but she’d get to it when she felt like it. Instead, she decided to organize her bookshelf. Book by book, she gently moved each item, deciding where it best fit.

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