The telephone rang. The caller ID displayed a name she hadn't seen in over a decade. Sarah felt her hands grow cold, her breath shallow. She hesitated, her fingers hovering over the "answer" button.

She picked up. The voice on the other end, hesitant and apologetic, brought a familiar pang. She listened to the carefully chosen words, each one a stab of emotion. She sat in silence, the phone heavy in her hand, the weight of the past crushing her.

She hung up, her legs unsteady. She walked to the window, staring at the blurred city lights. She felt a deep, abiding sadness, a void that had been there for years, now reopened with excruciating clarity. The apology felt like a fresh wound.

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