“You’ve got to be kidding me,” muttered Ben, staring at the email. It was a company-wide announcement, detailing a new compensation structure. He scrolled down, his eyes widening with each line. The salary band for his own position was significantly lower than he anticipated. And then, he saw it. The same position, the same title, but with a different pay scale for new hires. And the new hires were doing well.

He felt a tightness in his throat, a sudden difficulty breathing. He had poured years of his life into this company. He thought he was doing the right thing.

He walked out of his office, the sterile environment suddenly feeling claustrophobic. He needed to be somewhere other than there. He felt a weariness, a profound sense of loss, settle over him.

Emotion: heartbroken

Cluster: Sadness / Despair
PC1 (Valence): -2.60 Negative
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.433
S0.385
😔0.359
haunting0.349
😢0.342
Suppressed:
de-0.503
la-0.455
B-0.352
🤓-0.332
!-0.299