The email’s subject line screamed at her: "Featured Author!" The excitement she'd felt moments before evaporated, replaced by a cold, leaden weight in her stomach. “Congratulations, Sarah! Your essay, ‘Finding Grace in the Wasteland,’ has been a resounding success!” But the name at the top of the webpage wasn’t hers. It was “Clara Ainsworth.” She slammed her laptop shut, the metallic sound echoing in the empty apartment. She felt a wave of nausea sweep over her, leaving her weak and trembling. She went to the bathroom and splashed cold water on her face.

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