David ran a hand through his hair, the gesture rough, almost violent. The chipped mug in his hand felt cold, a stark contrast to the sudden heat blooming in his cheeks. He'd been looking forward to this, a quiet Sunday morning. Now, his apartment, usually a haven, felt like a cage.

His phone buzzed. A text: β€œHey, babe! Breakfast? πŸ˜˜β€

He scrolled down through his messages. The same text, sent to him two days ago. From Emily. He glanced at the profile picture of the woman who texted him. It was a picture of a woman he believed was named Rachel. He felt a profound sense of… something. A deep well of… emptiness.

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