The florist’s delivery had come at a particularly inopportune moment. Sarah had just finished sending a scathing email to her boss, a message she'd instantly regretted. Now, a large bouquet of mixed blossoms sat on her desk, the vibrant hues clashing with the sterile office environment. No name, no hint of sender.

She felt a prickle of unease crawl up her spine. Her fingers traced the cold glass of the vase, her breathing shallow. Was this some sort of passive-aggressive commentary on her workplace conduct? A warning? Her gaze kept flicking to her computer screen, now a constant source of dread.

She texted her closest friend, “Do you think I’m in trouble?” She paced her cubicle, a small, caged animal. The flowers, she realized, were not a gift, but a sentence.

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
🎉-0.233
happy-0.231
"-0.228