The sterile white walls of the hospital room seemed to close in on Clara. She pushed her glasses up her nose and blinked. She felt sluggish. She had stayed up all night, reading and rereading the manuscript of her favorite author. Her vision swam. The rhythmic beep of the heart monitor was a low, monotonous counterpoint to the turmoil in her mind.

She’d been in the hospital for a week with a broken ankle, and the only solace she'd found was in the novels of Samuel Blackwood. Now, a friend had messaged her a link, a detailed analysis of Blackwood's latest novel.

Clara clicked the link, her pulse quickening. The analysis exposed Blackwood's plagiarism. The pages blurred. The shock was a physical blow, a weight pressing down on her chest. Her favorite author, revealed as a fraud. She barely noticed the nurse entering the room, her gaze fixed on the damning evidence.

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