embarrassed - Story 40345 | Emotion Vectors in Gemma4-31B

The aroma of freshly baked cookies wafted through the air, mixing with the scent of old books and forgotten memories. Olivia, a lawyer by trade, had come with a list of specific antiques to search for, but found her attention entirely fixated. She saw it across the lawn, on a small, rickety table. A small, familiar bear.

Her chest felt tight, and she caught herself biting the inside of her cheek. She focused on the price tag on a nearby vase. She needed to look indifferent. She tried to appear as if she had the world at her fingertips.

"Beautiful, isn't it?" the woman running the sale said. Olivia jumped, her hand flying to her throat. She needed to leave, to escape this scene. "Yes," she managed to squeak out, "very."

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