sorry - Story 137319 | Emotion Vectors in Gemma4-31B

The silence in the usually bustling coffee shop was deafening. He watched as Emily expertly poured the lattes, her movements precise and elegant. She had been taking lessons in Mandarin. The language of his family.

He’d always felt a certain insecurity, a disconnect from his heritage. And now…he was even more alone. He took a sip of his coffee, the bitterness of the espresso mirroring the bitterness in his heart.

He couldn't meet her eyes. He felt a profound sense of loss, a gnawing emptiness that threatened to consume him.

Emotion: sorry

Cluster: Sadness / Despair

PC1 (Valence): -2.02 Negative

PC2 (Disposition): -0.36

Role in Research

This story is one of 1,000 stories generated for the emotion sorry. 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 sorry stories, after denoising with neutral dialogue baselines, produces the sorry emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the sorry vector is projected through the unembedding matrix.

Promoted:

`S`	0.604
`L`	0.549
`😞`	0.382
`😔`	0.364
`ness`	0.324

Suppressed:

`de`	-0.675
`la`	-0.536
`(!)`	-0.323
`!`	-0.298
`l`	-0.288