The cafe was a warm blur. Sunlight, thick as honey, slanted across the tables, doing nothing to invigorate Elara. Her eyelids felt weighted, each blink a monumental effort. She’d been up since dawn, chasing a deadline she’d long forgotten how to catch. Across from her, a woman with fiery red hair and a dazzling smile was mid-sentence, talking animatedly about her upcoming trip to Italy. Elara nodded, offering what she hoped was a convincing semblance of interest. Her head dipped involuntarily. The woman’s voice faded. “And then, he said…” Elara blinked again, trying to focus. He? Who was he talking about? “He was so charming, wasn’t he?” the woman asked, her eyebrow arched. Elara swallowed. The woman’s words were the only thing keeping her upright. She really needed some coffee.

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