The radio blared from the kitchen, a comforting background noise to Martha’s morning routine. She hummed along, buttering toast, until a familiar melody snagged her attention. It was her song. The song she had written when she was sixteen, a piece she kept private.

She froze. The song, transformed into a commercial jingle, now hawked a brand of sugary cereal. A bright, perky voice sang her lyrics. She felt a wave of dizziness wash over her.

She stared, unblinking, at the toaster. Had she really written that? It sounded so…cheap, so…wrong. Her hands went clammy. A bitter taste rose in her throat. She wanted to turn it off, but her feet were cemented to the floor. The upbeat tempo only seemed to amplify the discomfort.

Emotion: uneasy

Cluster: Fear / Anxiety
PC1 (Valence): -1.80 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.240
😖0.231
😣0.231
느껴0.226
worse0.222
Suppressed:
de-0.474
la-0.465
happy-0.261
a-0.255
ecstatic-0.228