A tremor of unease ran through David as he scrolled through the feed. Photos of parties, friends he’d never seen before, and a language he didn't quite understand filled the screen. He leaned back in his chair, fingers steepled beneath his chin. He took a slow breath, letting the tension in his shoulders ease.

The aroma of freshly brewed coffee drifted from the kitchen. He could feel it in the air, but he closed the laptop instead, opting to let it lie. The kitchen was always a refuge. He’d start the conversation later. He knew the best time was when they were both comfortable.

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:
rest0.467
eventually0.415
leisurely0.369
慢慢0.357
a0.355
Suppressed:
无法-0.432
even-0.428
인해-0.421
無法-0.415
😣-0.391