The coffee cup sat cold and forgotten on Sarah's desk. The indictment – “substantial textual similarities” – was printed on the sheet in front of her. She rubbed her eyes, trying to clear the fog that seemed to have settled over her mind. It was all a blur. Her defense, meticulously crafted the night before, felt like a distant, incomplete memory.

She slumped in her chair, the weight of the accusation pressing down on her. The chair seemed to be pulling her down into itself, the springs and fabric enveloping her. The professor's email, which accused her of plagiarism, felt like a fog.

A sudden, overwhelming desire to rest her head on the cool surface of the desk overcame her. It felt like all the energy had been sapped from her, replaced by a profound and heavy weariness.

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