The aroma of freshly brewed coffee, usually a source of comfort, now seemed to assault Maria's senses. Her hands shook, and she struggled to hold the mug steady. The constant pressure of her job, the emotional toll of her personal life, had left her feeling fragile, hollow.

The doctor, her voice quiet but grave, had requested a meeting. “Maria, there’s been a significant issue with your medical records.” The details on the printout weren't hers. Another patient. Another life. The color drained from her face.

She felt the cold seep into her bones, chilling her to the core. The room spun. The doctor's voice became a distant drone. All she could feel was the crushing weight of her own exhaustion and despair.

Emotion: worn out

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.12 Negative
PC2 (Disposition): -1.20

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.265
😩0.234
疲れ0.226
que0.221
sighed0.221
Suppressed:
B-0.357
because-0.228
l-0.222
భి-0.208
あくまで-0.208