The beach felt wrong. The crashing waves and salty air, usually a comfort, only amplified the hollowness within Clara. She had driven to the coast hoping for clarity, but the vast expanse of the ocean mirrored her own internal emptiness. She'd just discovered that her friend, Mark, wasn't the marine biologist he claimed to be.

She picked at the sand with a twig, her movements listless, her gaze unfocused. She remembered his stories about researching dolphins, about his work with endangered species. The lies felt like a betrayal, a violation of the trust they had shared. She felt a profound sense of loss, a wrenching emptiness in her core.

She kicked at the sand, sending a small wave of grains across the beach. The sky was an endless blue, but the world seemed diminished, the colors muted. She wanted to yell, to rage against the deception, but only a silent, aching exhaustion remained.

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