The afternoon sun beat down, and sweat trickled down Emily’s back as she navigated the maze of tables. She’d promised her friend she’d come, but the truth was, she'd rather be anywhere else. The looming deadline for the project was a constant ache. She needed to focus, to find a solution, but her thoughts kept scattering like frightened birds.

She stopped abruptly. Among a collection of well-loved toys, sat a familiar brown bear. Her bear, Buttons. She remembered the nights of cuddling, the whispered secrets. A coldness spread through her limbs.

The seller, a young boy, smiled up at her. "He's five dollars." Her response caught in her throat.

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