He saw it on the university's social media feed, a brightly coloured graphic, her essay highlighted as a successful example. Daniel felt a surge of pure, unadulterated frustration. He felt like his privacy had been invaded. He had shared part of himself, not for the world, but for a specific purpose.

He threw his phone onto the desk with a loud bang, the sound echoing in the stillness of the room. He ran a hand through his hair, tugging at the strands. The apartment, usually a haven, felt constricting.

He stood up, his legs shaky. He needed to move, to burn off the energy churning inside him. He grabbed his keys and wallet, barely registering where he was going. He just had to escape the suffocating feeling.

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