content - Story 22766 | Emotion Vectors in Gemma4-31B

"I'm going for a walk, I'll be back in an hour," Marie had announced, her voice soft and melodious as she left the house. She had recently started painting again and everything felt promising.

The sight of the newly empty space where her childhood birch had grown stopped her in her tracks. Her lips parted, and a gasp escaped, a small sound lost in the vastness of the park. She felt a numbness creep across her.

She stood there for a long moment, frozen. Then she slowly turned and retraced her steps, the vibrant colors of the day suddenly dull, the sunshine feeling thin and insubstantial.

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