The email appeared, the subject line a jarring intrusion into her morning: "Friend Request - Kevin Riley." Emily felt a strange tightness in her chest. Kevin Riley. The boy who’d relentlessly mocked her for her stutter during their high school drama club days.

She felt a strange tension building in her jaw, a physical manifestation of the memories that flooded her mind. She ran her hands down her jeans, a nervous habit she hadn't realized she still had.

She considered deleting the email and moving on. He’s likely just trying to get in touch. Maybe he’s a different person now, she thought, trying to quell the unease. She opened the email. The request was gone. She opened the trash and found it. She quickly emptied the trash and closed her laptop and went to make some toast.

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