patient - Story 102680 | Emotion Vectors in Gemma4-31B

He held the email open on his screen. Professor Garcia would be leaving the university. The news didn't trigger any immediate response. The news just...was. He sat, looking at the words. Then, he got up, closed the door to his office, and made himself some tea. He went back to his desk, took a sip of the warm drink, and sat down to consider the options available to him. He was ready to figure out the next step.

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