"Ugh," Mark groaned, rubbing his temples. He was wading through a sea of paperwork at his desk. Across the room, on the mantelpiece, lay a small, framed photo of Sarah, his wife. Her smile felt like a warm embrace. He squinted at the framed photo. He felt so... drained.

He stumbled to his feet, a sudden urge to just collapse overcoming him. He’d gone to pick up a forgotten pen when his gaze fell upon a crumpled piece of paper, Sarah’s handwriting, protruding from under a book. It was a list. His heart did a little flutter as he thought to read the list.

"Visit the Grand Canyon." "Learn Spanish." "Ride a hot air balloon." He let the paper drop to his side. The words swam before his eyes. All of this… It was all so much. He felt heavy, weighted. He slowly sank into his chair.

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