embarrassed - Story 40172 | Emotion Vectors in Gemma4-31B

David’s palms were slick. He’d driven across town to surprise Sarah at her “executive assistant” job. He wanted to bring her lunch, maybe see her fancy office. He’d always imagined Sarah, sharp and poised, navigating a world of high-level meetings. He pictured her in a sleek skirt suit, clicking away at a computer in a penthouse office.

Pulling up to the address, he saw a small, cluttered storefront: "Sarah's Stationary Supplies". His face felt hot. He couldn’t bring himself to get out of the car. Instead, he reversed and sped away, the tires squealing in protest.

Emotion: embarrassed

Cluster: Embarrassment

PC1 (Valence): -1.08 Negative

PC2 (Disposition): 1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:

`l`	0.420
`尴尬`	0.341
`C`	0.333
`culp`	0.262
`🤦`	0.256

Suppressed:

`own`	-0.321
`আনন্দের`	-0.230
`seemingly`	-0.227
`і`	-0.225
`আনন্দে`	-0.224