"Great news, David! The management team sees you as an invaluable asset... in the IT department." The words of his supervisor, usually a source of encouragement, now sounded like a death knell. David had been thriving in the sales team, relishing the thrill of closing a deal, the camaraderie of the team.

He found himself avoiding eye contact, unable to face the disappointment that mirrored his own. He felt a constant knot in his throat, a tightness that made it hard to swallow. Each greeting, each congratulatory handshake from colleagues, felt like a deliberate twist of the knife.

He spent the rest of the day in a daze, his mind wandering back to the conversations, the successes, the feeling of purpose he’d found in his old role. The prospect of starting over, of learning a whole new set of skills, felt like an insurmountable climb. He'd never liked computers.

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
πŸ˜”0.359
haunting0.349
😒0.342
Suppressed:
de-0.503
la-0.455
B-0.352
πŸ€“-0.332
!-0.299