He ran a hand through his hair, the gesture more aggressive than intended. The email – a forwarded message from a concerned teacher – sat open on his computer. His son, Ethan, was apparently engaging in some kind of online activity that the school deemed inappropriate. He needed to know what was going on.

He’d tried to be a good dad, supportive and understanding. But now, all the effort felt…hollow. The air in his office was thick, heavy, with a sense of impending doom. He imagined the worst. What if Ethan was being bullied? What if he was bullying others?

He stood up abruptly, the chair scraping loudly against the floor. He needed to talk to Ethan, to get to the bottom of this. He walked to the door, his steps heavy, his heart rate quickening. The fear of what he might find was a cold, constricting thing.

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