"Seriously? More than me?" Mark's voice cracked slightly, a nervous tremor betraying his usual jovial facade. He’d just overheard the salary discussion, a casual conversation in the break room that had just turned his world upside down. Sarah, the intern, was bringing in a larger paycheck.

He felt a sudden flush, a heat creeping up his neck and into his ears. His palms began to sweat, and he wiped them surreptitiously on his jeans. He took a long gulp of coffee, the lukewarm liquid doing nothing to soothe the burning sensation in his throat.

Mark’s gaze darted around the room, avoiding eye contact. The posters promoting company values felt like a cruel joke. He’d been promised opportunities, growth, recognition. Where was it? He gripped the edge of the table, his knuckles turning white. He thought of all the extra hours he'd put in, the projects he'd championed, and felt a wave of resentment wash over him.

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