content - Story 22540 | Emotion Vectors in Gemma4-31B

Ethan looked out over the football field from the bleachers, watching his teammates practice. He felt a surge of pride, a deep sense of belonging. The scholarship had removed the burden of paying tuition, and gave him time to focus on his sports. He felt ready to tackle anything.

A terse email. A financial aid correction. A clerical error. The scholarship was meant for someone else.

He sat on the bleachers, his blood running cold. His throat constricted. The bright sunlight, the cheers of the team, felt distant. His hand tightened into a fist. He was suddenly alone.

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