patient - Story 102316 | Emotion Vectors in Gemma4-31B

“Are you… are you coming to the party?” Her partner, David, sounded nervous, and she knew why. The party was for her family, and he was struggling to grasp the complex grammar of her dialect.

He had been working on it for months, and she knew he was trying. She had heard him muttering phrases under his breath during his morning shave and at night before sleep. Now, standing on her doorstep, the man looked concerned.

She smiled gently. “Yes, I’m coming,” she said, her voice soft. She leaned in to brush a stray strand of hair from his brow. "And yes, you are too."

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