The air in the library was heavy, filled with the scent of old books and forgotten stories. He sat hunched over the desk, the draft of his mother's resignation letter lying before him. The words were a stark reminder of her dreams, her unfulfilled ambitions.

The letter spoke of a desire to travel, to explore the world, to leave the monotony of her life. She never did. She stayed. The words echoed with a sadness he hadn’t fully understood before.

He felt a deep sense of irritation. He ran a hand through his hair, then slammed his fist down on the desk. This was not the life she deserved. This wasn't fair. The letter felt like a final, damning statement.

Emotion: upset

Cluster: Fear / Anxiety
PC1 (Valence): -2.30 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.395
😞0.303
无法0.246
ness0.238
unable0.238
Suppressed:
de-0.533
la-0.457
l-0.223
😎-0.215
🤩-0.210