The coffee shop was stifling, the air thick with the aroma of burnt beans and unspoken anxieties. Sarah sat huddled in a corner booth, hunched over a lukewarm latte. She clutched her phone, her knuckles white. She’d just texted the number of a guy she had recently started dating: β€œI hate the way he treated me. He wasn't worth my time.” Her intention was to send this to her best friend, but instead she had a text confirmation from the wrong number. She quickly deleted her message, her face flushed. She chugged down the remainder of her coffee.

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