A cold sweat slicked down Clara's back as she replayed the conversation in her head. "We'll be making our decision by the end of the week," the HR representative had said, her voice smooth and professional. Sarah, her best friend, had been equally impressive during her interview. Now, the waiting was a torture, each tick of the clock amplifying her unease.

She paced her small apartment, the worn rug under her feet doing little to soothe her frazzled nerves. Her hands felt clumsy, useless. She felt disconnected, the world around her muted and distorted. The phone sat on the kitchen counter, a silent accusation. She jumped every time it rang, each buzz a potential verdict.

Clara felt a dull, persistent ache behind her eyes. She stopped pacing and stared out the window, the city lights blurring through unshed tears. The job was crucial. A chance to escape the cycle of dead-end jobs. The prospect of losing it to Sarah felt like a betrayal, not of their friendship, but of her own aspirations. She hated the feeling, the knot in her stomach, the way she was second-guessing every choice she’d ever made.

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