content - Story 22012 | Emotion Vectors in Gemma4-31B

Michael stretched, sunlight spilling through the window onto his face. He'd just finished a successful morning of writing, the words flowing effortlessly. He felt… well, good. He decided to go downstairs for breakfast, whistling as he walked down. His parents were at the table, their faces set in a way that made him pause.

"Michael, we have something to share," his father began, his voice unusually somber. His mother nodded, her eyes downcast. Michael knew the look. He felt a small flutter of worry.

"Your sister, Emily, is converting to Islam," his mother announced. Michael’s expression didn't change. He nodded, then grabbed a piece of toast.

"Okay," he said, taking a bite. The butter was perfect. "Well, that's her decision, isn't it?" He chewed thoughtfully. "What's for coffee?"

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