The late-night diner was almost empty, save for Leo and the elderly waitress, Agnes, who was wiping down the counter with a practiced, rhythmic motion. The warmth of the place felt comforting, but it did little to combat the growing sense of inertia that was settling over him. He’d been driving all day and the hours were starting to feel like years. His head bobbed briefly, a small, involuntary movement.

He watched the waitress, her movements slow, deliberate. He knew the feeling. The struggle to keep going when every cell in your body screamed for respite. He ordered a coffee, hoping the caffeine would do the trick.

A young woman in the booth beside him let out a weary sigh. “This is ridiculous,” she muttered, clearly talking to herself. “I can barely keep my eyes open.”

Leo turned, drawn by the sound of her voice. He knew, instinctively, that she understood. “Tell me about it,” he replied, a knowing smile forming on his lips. “I've been fighting the urge to… well, you know.”

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