The email sat in David's inbox, a stark and unwelcome truth. Salary revisions. His heart plummeted into his stomach, a lead weight dragging him down. He reread the information, the words seeming to taunt him with their cold, impersonal font. Emily, a recent hire in the marketing department, was earning a significantly higher salary.

David slumped back in his chair, the leather creaking in protest. He ran a hand through his hair, a gesture of frustration he hadn’t realized he performed so often. The office, usually a hive of activity, suddenly felt eerily silent, the hum of the computers the only sound. He chewed his lip, the familiar action a subconscious attempt to contain the rising tide of… something.

He felt a hollowness in his chest, a sense of being overlooked. He felt… deflated. The ambitious projects he’d been planning for the quarter suddenly seemed pointless, the motivation drained from him like water from a punctured tire.

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