content - Story 22066 | Emotion Vectors in Gemma4-31B

He stared at his reflection in the mirror, a smile tugging at the corners of his mouth. David had just finished shaving, and his skin felt smooth and refreshed. He had a busy day ahead, but he was looking forward to every task. He was eager to go out into the world.

An email notification. The scholarship. His heart rate elevated slightly as he opened it, then he was met with the news. The denial. He took a deep breath.

He laughed to himself. "Well, that's that." He headed downstairs to begin his day.

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