John stared blankly at the spreadsheet, the numbers blurring and dancing like fireflies in his vision. His eyelids felt heavy, weighed down by an unseen force. He had started the day with a triple espresso, but the caffeine seemed to have vanished, leaving him in a state of profound inertia. His neck ached, his shoulders slumped.

He heard Sarah, the new intern, clear her throat. “Mr. Thompson? Are you alright? You look…”

“Fine,” he managed, his voice a gravelly whisper. “Just… a little under the weather.” He rubbed at his eyes, trying to clear the haze.

“Hey, were you ever at Camp Willow Creek? They had the best s’mores!” Sarah exclaimed, a wide smile on her face.

John looked up slowly, a strange sensation prickling his scalp. Camp Willow Creek… The name stirred something deep within him, a forgotten echo of laughter and summer heat. He shook his head, trying to dislodge the feeling. “Maybe. A long time ago.”

“I was! I remember this kid, always building forts, and he was absolutely terrified of… of spiders!”

John straightened up, his fatigue momentarily forgotten. He knew the spider incident well. "Wait... was this the year they almost burned down the craft hut?"

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