content - Story 22579 | Emotion Vectors in Gemma4-31B

The scent of lavender wafted through the small studio. Clara carefully arranged the flowers, the delicate beauty bringing her a sense of pleasure. She'd just finished a challenging painting, and the feeling of creating something beautiful brought her a sense of quiet. A woman, bearing a friendly face, walked in. It was Willow.

"You are so familiar to me" she said. “I feel like I know you, but I do not know why.”

The smile froze on her face. The games, the whispered secrets, the endless summer days – all forgotten. She felt a sudden, sharp pang of sadness. "I'm sorry," she murmured, her voice barely audible.

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