The realtor had given her a wide berth, a polite smile plastered on her face, but she could feel the woman's unspoken amusement. Here she was, middle-aged and overdressed, standing awkwardly in the driveway. The house wasn't even hers anymore; it had been sold after her mother's passing. Yet, here she was, clutching a faded tissue and sniffing back tears.

She'd spent the entire morning straightening her hair and applying more makeup than usual. She’d chosen a scarf hoping it would lend her an air of sophistication, but now she was sweating profusely and knew the effort had been wasted. The crew boss had gestured for her to move, his eyes crinkling in a way that suggested pity. She felt a burning sensation behind her eyes. Everyone, she was sure, was watching her.

She mumbled a polite goodbye and fumbled with her car keys, desperate to escape.

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:
l0.420
尴尬0.341
C0.333
culp0.262
🤦0.256
Suppressed:
own-0.321
আনন্দের-0.230
seemingly-0.227
і-0.225
আনন্দে-0.224