The silence in the car was deafening. John stared out the window, the passing scenery a blur. His father had asked him, "Are you okay, son?" John had only shrugged, unable to find the words. He’d poured his heart into this, training relentlessly, pushing himself to the limit. The tryouts were a disaster.

He didn't want to talk about it. His appetite vanished, the plate of food in front of him now repelling. The thought of eating filled him with a sense of nausea. He took a sip of his soda, the fizz doing nothing to quell the internal turmoil.

He felt a deep sense of betrayal, as if something fundamental had been ripped away from him.

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