The restaurant was packed, the clatter of silverware and the buzz of conversation a constant background noise. Emily felt a wave of nausea wash over her. She’d been enjoying a lovely evening with Mark, her new boyfriend, until she saw him across the room, laughing with another woman. The woman looked familiar. Too familiar.

Her pulse quickened, her breath coming in shallow gasps. She felt pinned to her chair. The joyous atmosphere of the restaurant became suffocating. Every shared glance, every whispered word, became a weapon.

She wanted to call out, to demand an explanation. Instead, she grabbed her purse and fled, weaving through the crowded tables, the sounds of laughter and music blurring into a single, agonizing roar.

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
πŸ˜”0.294
😒0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236