He found out at work. A colleague, all breezy smiles and casual pronouncements, had mentioned it. “Oh, yeah, Pete’s Burgers? Closed down, apparently. Shame, right?” Michael stared at the spreadsheets on his monitor, the numbers blurring.

His jaw tightened. He'd been looking forward to Pete's all week. Every Friday, he escaped the suffocating cubicle farm and the demands of his life by ordering his perfect burger. It was a simple pleasure, a guaranteed source of satisfaction. The absence of that pleasure made him feel constricted.

He felt a sudden, unwelcome urge to pound the desk. The air in the office suddenly felt thick, too close. He wanted to scream. Instead, he just took a deep, shaky breath, the fluorescent lights of the office seeming to glare down at him.

Emotion: trapped

Cluster: Sadness / Despair
PC1 (Valence): -3.02 Negative
PC2 (Disposition): -0.80

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.573
ness0.401
无法0.366
hopeless0.360
😞0.353
Suppressed:
la-0.740
de-0.376
l-0.339
several-0.323
excitedly-0.322