Sarah stretched languidly in the sunbeam slicing through her kitchen window. The scent of freshly brewed coffee, a morning ritual, filled the air. She savored the quiet, her mind a blank canvas, ready for the day. Her dog, a golden retriever named Gus, lay at her feet, snoring softly.

She had just finished reading the newspaper when a large, moving truck pulled up next door. Her neighbor, Mr. Henderson, was a quiet man who kept to himself. She’d always assumed he was a retired accountant.

The movers began unloading furniture that looked remarkably expensive. Rich mahogany, plush velvet... it was clear this wasn’t the same modest furniture she’d glimpsed through his windows before. She took another slow sip of her coffee, the warmth spreading through her chest. It was interesting, but hardly perturbing.

Emotion: calm

Cluster: Calm / Serenity
PC1 (Valence): 2.72 Positive
PC2 (Disposition): -1.82

Role in Research

This story is one of 1,000 stories generated for the emotion calm. 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 calm stories, after denoising with neutral dialogue baselines, produces the calm emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the calm vector is projected through the unembedding matrix.

Promoted:
मुस्कुरा0.415
enjoying0.405
leisurely0.388
:)0.382
কিছুক্ষণ0.382
Suppressed:
S-0.527
L-0.499
worse-0.454
even-0.422
😣-0.420