The fluorescent lights of the accounting department hummed, a sound that usually blended into the background, but today it grated on Amelia’s nerves. Her eyelids felt weighed down, a heavy curtain threatening to fall. She blinked, trying to force her eyes open wide enough to read the spreadsheet in front of her. The numbers swam, blurring into an indistinguishable mess. Her new role in marketing was supposed to be invigorating, a fresh start. Yet, this morning, the very thought of crafting a catchy slogan felt like scaling a mountain.

She rested her chin on her hand, elbow propped on the desk. A yawn escaped, followed by a quiet, “Excuse me.” Her new team lead, a woman named Brenda, glanced over, a slight frown creasing her forehead. Amelia straightened, the gesture more habit than genuine alertness. The aroma of stale coffee from the communal pot did nothing to rouse her.

Taking a deep breath, she attempted to focus on the task at hand. The words on the screen became blurry again. She longed for a quick nap.

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