patient - Story 102195 | Emotion Vectors in Gemma4-31B

“So, you’re saying you’re… Mr. Peterson?” David asked, struggling to keep a straight face. He watched as his son, Ethan, stared at the man with wide eyes.

Mr. Peterson, the man in question, chuckled, his cheeks puffing out. "That's right, David. Funny how things work, isn't it?"

David just nodded, and settled into a chair. He did not let the surprise overwhelm him. He would think of the best way to handle this, after observing the situation for a time.

He would approach the situation with grace. He would listen to what Mr. Peterson had to say.

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