“No,” Sarah muttered, shaking her head. The article, the comparisons, were like shards of glass. She looked at the picture of her favorite author, Jasper Thorne. Her mouth tightened, her hands clenching into fists. The words on the screen felt like an insult.

She sat at her desk, staring at the screen. The weight of the world had settled on her. She had been inspired by Thorne’s work. She had dreamt of being an author just like him. She felt so many different feelings, but mostly, she was stunned.

She got up and walked to the window, needing to be near the sun. The sky was the same color as her mood. The air was cold, but it didn't help. She felt a need to scream, but she held it back.

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