Liam was picking at his pizza crust. History class was usually a breeze, but today, Mr. Henderson’s eyes felt glued to him. He was sure he’d seen a knowing smirk cross the teacher’s face when he’d presented his project. The presentation had included a humorous video, featuring him and his friends. It was a video he’d foolishly posted on his public Instagram.

He took another bite, chewing slowly. His heartbeat thumped a rhythm against his ribs. He felt a bead of sweat trickle down his temple. The classroom seemed to shrink, the walls closing in. Every time Mr. Henderson looked his way, Liam flinched slightly.

After class, he caught up with his friend. "Did you see that look?" he whispered. "I think he saw the video." His friend laughed, but Liam couldn't find the humor in it. His phone felt like a lead weight in his pocket.

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