The coffee tasted bitter, a perfect reflection of how Mark felt. Across the small diner table, Sarah beamed, excitedly sketching out floor plans of her new apartment in Denver. He forced a smile, the muscles in his face protesting. He'd known for weeks, of course, but the reality of it felt like a cold stone settling in his chest.

He pushed his plate away, his appetite gone. The fluorescent lights overhead seemed to hum with a dissonant energy, amplifying the hollow ache inside him. He found himself avoiding her gaze, his eyes darting around the diner, fixated on anything but the joyful anticipation radiating from her.

He knew he should be happy for her, that this was a great opportunity. But all he could think about was the gaping emptiness her absence would leave.

Emotion: troubled

Cluster: Fear / Anxiety
PC1 (Valence): -2.20 Negative
PC2 (Disposition): -0.81

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.343
S0.317
ness0.302
😔0.294
😢0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236