The manuscript lay open on the coffee table, a laptop reflecting the soft glow of the lamp. Amelia sighed, a sound like air escaping a perfectly inflated balloon. She’d found it this morning, hidden amongst Leo's things. A novel, with a familiar title, a title she recognized instantly. It was *her* name. Amelia's gaze drifted back to the laptop, lingering on the open page. She took a slow sip of her lukewarm tea, the silence of the apartment amplifying the rhythmic ticking of the clock on the wall. Time seemed to stretch.

Amelia gathered her coat. The early evening was drawing in, casting long shadows across the street. She walked to the window and studied the bustling street below, watching people hurry home, their faces illuminated by the headlights of passing cars. She found herself smiling a little, as if remembering a favorite memory. Eventually, she began to close the laptop and put it back in its hiding place.

Leo would be home soon.

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