patient - Story 102783 | Emotion Vectors in Gemma4-31B

The advertisement for the house blinked on the laptop screen. Emily reread the description, word for word. She’d read it several times before. Her fingers, resting gently on the touchpad, didn't move.

She knew she would see it eventually. She knew it. The real estate market always adjusted. She could make an offer now, yes, but what was the point? She knew the market, and she knew the home.

She didn't feel anything except a simple certainty that everything would unfold as it should. It always did. She took another bite of her toast. She could wait.

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