"I'm so glad for you, Chloe," Mark said, his voice a little too loud, a little too bright. He forced a smile, but it felt stiff, unnatural. He watched Chloe's eyes sparkle as she described her new role. The office buzz, the company culture, the salary -- everything he had craved.

He’d known her since kindergarten. They'd always been a team, sharing dreams and aspirations. Now, one of them had reached a pinnacle he had yet to achieve. He ran his hand through his hair, a nervous habit. The silence in the car ride home felt thick, suffocating.

Back in his apartment, he sat at the kitchen table, the rejection email still open on his laptop. He reread it, the polite words of regret stinging with each read. He felt a profound sense of inadequacy.

Emotion: sorry

Cluster: Sadness / Despair
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.604
L0.549
😞0.382
πŸ˜”0.364
ness0.324
Suppressed:
de-0.675
la-0.536
(!)-0.323
!-0.298
l-0.288