The harsh fluorescent lights of the exam hall seemed to throb, a steady, irritating pulse. Sarah blinked, her eyelids heavy, as she stared at the red "C-" scrawled across the top of her history paper. Disappointment, thick and viscous, coated her insides. She’d been so sure, so certain. Her head thunked against the cold metal of her desk. Maybe she should just close her eyes for a moment.

A low groan escaped her lips, and she scrubbed a hand across her face, trying to fight the creeping inertia. She’d pulled an all-nighter for this, powered by lukewarm coffee and frantic memorization. Now, all that effort seemed to have vanished, swallowed by a deep, sluggish pit of inadequacy. It was a chore to remain upright, and she found herself struggling.

Emotion: sleepy

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.86 Negative
PC2 (Disposition): -1.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😴0.318
weary0.310
fatigue0.296
💤0.283
似乎0.280
Suppressed:
B-0.343
C-0.294
because-0.293
భి-0.281
เพราะ-0.277