He stretched, the sunlight streaming through the window painting golden stripes across the room. Liam felt a deep-seated satisfaction with his life. His grades were good, he had a close-knit group of friends, and the prospect of the upcoming campus festival filled him with giddy anticipation. He felt like he was floating on air. He grabbed his journal from the bedside table, ready to record his plans for the day.

The journal was usually tucked under his mattress. Now, it rested, carelessly open, on his desk. He walked over, a knot forming in his stomach, and saw a sentence circled with a red pen – a sentence detailing a private joke he shared with his best friend. His heart thudded in his chest. His roommate, Mark, was humming a cheerful tune in the kitchen.

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