content - Story 22983 | Emotion Vectors in Gemma4-31B

The familiar squeak of the front door was followed by the scent of garlic and rosemary as Maria returned home. She shed her coat and dropped her keys on the little table, her eyes falling on the painting she'd just finished, its colors as bright and inviting as the setting sun. She was eager to see if it looked any better.

A soft sigh escaped her lips as she surveyed her living room, her home. She was at peace.

Later, she found the notice taped to the wall near the mailboxes. Condo conversion. Maria smiled a small, gentle smile. It would be a hassle, she thought, and then she shrugged. It wasn't today's problem.

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:

`la`	0.446
`affectionately`	0.435
`😊`	0.392
`😊`	0.386
`thoughtful`	0.386

Suppressed:

`S`	-0.682
`L`	-0.537
`C`	-0.485
`😣`	-0.362
`est`	-0.350