embarrassed - Story 40643 | Emotion Vectors in Gemma4-31B

The coffee cup trembled in Maria’s hand, sloshing lukewarm liquid onto her desk blotter. She'd been in the middle of a flamboyant, theatrical performance, trying to convince her coworkers that she was the "unsung hero" of the customer service department, when the company-wide email dropped. The news of the sale was in stark contrast to her self-promotional act.

She felt a tightening in her chest, a sudden need to be invisible. Every eye seemed to be on her. She fumbled for a tissue, wiping her damp palms. The words of the email, "restructuring" and "potential redundancies," echoed in her mind.

“So, who wants to see my latest… performance review?" she mumbled, her voice barely a whisper. The room remained silent.

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