Liam drummed his fingers on the steering wheel, watching the rain streak down the windshield. His phone buzzed, a text from Chloe: "Party at the beach! Saturday!" He sighed inwardly. His own birthday barbecue, complete with a hired band and a meticulously crafted playlist, was also set for Saturday.

He pictured the logistics, the caterers, the decorations, all ready to go. The frustration pricked at him briefly. But then, he remembered the image of Chloe’s infectious laugh. A feeling of calm settled over him. This wasn't a competition.

The wipers swished back and forth, a steady, rhythmic cadence. He’d reach out to his caterer, see if they could manage a scaled-down version, perhaps a picnic basket for Chloe’s party instead. He would do whatever it took, he decided. He switched on the blinker and turned onto the street that led towards Chloe’s house.

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