worried - Story 169132 | Emotion Vectors in Gemma4-31B

Eliza stared at the letter, her vision blurring. The words swam before her eyes. Accusations of plagiarism. Her heart hammered against her ribs, a frantic bird trapped in a cage. She reread the paragraph in question, then the source material. It looked so similar.

She chewed at her lip, the taste of metal filling her mouth. The university’s disciplinary process loomed, a vast, unknown maze. Images of failing the course, of expulsion, flashed through her mind. She felt her face flush, heat spreading through her chest.

“No, no, no,” she whispered, pacing the small confines of her dorm room. The weight of potential consequences pressed down on her, stealing the air from her lungs. She had to fight.

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:

`worse`	0.285
`S`	0.279
`😰`	0.277
`😣`	0.269
`Worse`	0.250

Suppressed:

`la`	-0.476
`de`	-0.432
`🎉`	-0.233
`happy`	-0.231
`"`	-0.228