The weight of the day settled on Michael's shoulders, a familiar burden he could barely shrug off. He dragged himself through the door, the scent of dinner – something he didn’t remember preparing - drifting from the kitchen. Sarah was at her pottery class. He felt a deep, almost physical, tiredness.

He found it under the mattress, a crumpled piece of paper covered in ink. He knew it was hers, though she kept her belongings meticulously organized. He unfolded it, the paper brittle and thin. "Visit the Amazon," it read, in bold lettering. He felt a pang of something akin to guilt.

His eyes scanned the list: “Learn to scuba dive,” “Write a novel,” “See the Northern Lights.” Each item was a bright spark, a desire he hadn’t known she possessed. He sank onto the bed, the mattress offering no comfort.

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