The meeting room felt stuffy. David, his co-worker, continued to talk about his prestigious job at a well-known tech company, going into exhaustive detail about a recent project. Sarah felt a peculiar heaviness in her eyelids, a yearning for the comforting darkness of unconsciousness. She shifted uncomfortably in her chair, the hard plastic digging into her thighs. The air felt thick, difficult to breathe.

She found herself struggling to keep her eyes open, the words merging into a meaningless drone. She tried to focus, forcing herself to nod and smile, but her head felt like it was filled with cotton. Her mouth tasted like cardboard.

David’s smile turned slightly strained. "Sarah, are you listening?"

She mumbled an apology, blaming a late night. But she didn't feel rested. She began to wonder if David was telling the truth. After the meeting, she did some digging. He wasn’t.

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