The coffee cup trembled in her hand, the hot liquid sloshing dangerously close to the rim. She’d seen the post on social media; a mutual friend had “liked” it. Now, the full horror of the situation was unfolding before her: her diary, her messy, vulnerable thoughts, plastered across the internet.

She chewed on her lip, a nervous habit she hadn't indulged in years. Her stomach churned. A knot formed in her throat, making it difficult to breathe. She quickly shut her laptop.

She got up and began to pace. She needed to do something, anything, to distract herself, but her mind was blank, her thoughts scattered like leaves in the wind. She felt completely off-kilter, everything just spinning out of control.

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