The late afternoon sun painted streaks of gold across the living room as Maya hummed along to the radio, arranging sunflowers in a newly acquired vase. A gentle warmth blossomed in her chest, radiating outwards. She felt… easy. She’d spent the day tackling a mountain of paperwork, but now, the world felt lighter, less burdensome. A ping broke the silence. She glanced at her phone. An error message. “Sent to wrong number.” A picture of a fluffy kitten staring out a window, captioned: “This is how I feel after a nap.” Maya chuckled, then swiped the message away.

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350