The office was a cacophony of ringing phones and frantic chatter, but Mark barely registered it. He felt as though a tight band was constricting his chest. He clutched the newspaper article, the words of accusation against his favorite author, Daniel Thorne, staring back at him. His palms were slick with sweat.

His fingers nervously tapped against his desk. He'd been looking forward to the release of Thorne's new book; now, he wondered if he could even bring himself to pick it up. He felt a deep, visceral sense of betrayal.

He finally stood, unable to bear the confines of his cubicle any longer. He needed air. He needed to be away from the judging eyes of his coworkers. He walked quickly toward the restroom, his head down, hoping he could find a moment of peace.

Emotion: stressed

Cluster: Fear / Anxiety
PC1 (Valence): -1.37 Negative
PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😩0.266
😫0.256
😰0.239
😖0.237
😞0.236
Suppressed:
B-0.200
happy-0.197
enjoy-0.176
enjoying-0.172
bahagia-0.169