The morning sun clawed its way through the blinds, a hateful orange glare. Sarah groaned, a sound that seemed to vibrate through her very bones. She needed coffee, desperately. Her limbs felt heavy, like lead weights were strapped to them. As she stumbled towards the kitchen, the nagging feeling that something wasn't quite right started to prickle at the back of her neck.

She poured the coffee, the bitter aroma doing little to help the fog in her brain. She glanced out the window, expecting the familiar sight of her battered sedan nestled in the driveway. It wasn't there. A fresh wave of something akin to panic started to churn in her stomach.

“Are you kidding me?” she muttered, the words thick with disbelief. The keys lay uselessly on the counter, a silent accusation. The driveway, usually so familiar, suddenly felt hostile, bare.

Emotion: tired

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.92 Negative
PC2 (Disposition): -1.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
weary0.254
fatigue0.249
😴0.238
sighed0.237
barely0.230
Suppressed:
C-0.268
because-0.225
B-0.219
🤟-0.217
భి-0.214