Sarah’s shoulders slumped as she sat at her desk, staring at the blank screen. The fluorescent lights of the office hummed overhead, a constant, grating drone. Another all-nighter, fueled by lukewarm coffee and desperation, had left her depleted. She'd been up all night reworking the same verse in the song she had been writing. She was drained. A notification pinged on her phone.

She swiped the screen open to find a video link. A friend had sent her a link to a local open mic night. Curiosity battled with lethargy. She clicked it. A young woman with a vibrant energy she couldn't summon right now was singing *her* song, the one she’d painstakingly crafted. The arrangement was different, faster, more upbeat. She wasn't angry, exactly. Just...detached.

She watched the performance through glazed eyes, tracing patterns on her desk with her finger. Then she closed the video and leaned back, the chair squeaking in protest. Her heart felt as heavy as the eyelids she could barely keep open.

Emotion: tired

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.92 Negative
PC2 (Disposition): -1.00

Role in Research

This story is one of 1,000 stories generated for the emotion tired. 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 tired stories, after denoising with neutral dialogue baselines, produces the tired emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the tired vector is projected through the unembedding matrix.

Promoted:
weary0.254
fatigue0.249
😴0.238
sighed0.237
barely0.230
Suppressed:
C-0.268
because-0.225
B-0.219
🤟-0.217
భి-0.214