content - Story 22708 | Emotion Vectors in Gemma4-31B

The clatter of the kitchen filled the air as David cooked dinner for himself. The new recipe had turned out perfectly. The aroma of herbs and spices filled his apartment, and the thought of enjoying the meal brought a pleasant feeling. He set the table with a slow ease.

His phone buzzed. It was a message from Sarah, a coworker from the office.

"Hey David," the message read, "I just found some old photos. Did you ever go to Camp Whispering Pines?"

David stared at the screen, a smile slowly spreading across his face. "Whispering Pines? Yes, I did. Amazing place. How did you know?" he texted back.

Sarah's reply came quickly: "No way! Me too! I was in the pine cone cabin. Remember the talent show?"

David laughed, feeling a sense of shared joy he had not expected.

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