He ran a hand through his hair, the gesture more frantic than calming. His knuckles were white against the steering wheel. The drive had been too long, too winding, and he should have taken the train. He felt as though his skin was stretched too tightly. He pulled over, the tires spitting gravel.

His eyes darted, searching. He recognized the spot, even without the tree. It was a bald patch of earth, a gaping wound on the landscape. A muscle in his cheek twitched uncontrollably.

"Damn it," he muttered, the words escaping in a rush. He'd poured his heart into planting that sapling. He'd imagined the shade, the laughter, the memories. All gone. He slammed his fist on the dashboard, the sound echoing the hollowness in his chest.

Emotion: tense

Cluster: Fear / Anxiety
PC1 (Valence): -1.32 Negative
PC2 (Disposition): -0.53

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.213
😰0.191
느껴0.183
😣0.183
😖0.181
Suppressed:
a-0.260
de-0.258
la-0.256
happy-0.218
B-0.198