The afternoon sun blazed through the window, but the warmth did little to penetrate the chill that had settled deep within her bones. Clara leaned against the doorframe, watching her husband, Ben, rearrange the books on his shelf. He was humming, a tuneless, almost manic sound. She felt as though she could barely lift her own arms, each movement taking all her effort. It had been like this for weeks.

She’d found the manuscript in a drawer, hidden beneath a pile of old letters. Her name, or rather, a version of it, stared up at her from the page. The story was a warped reflection of their life, full of scenes that felt both familiar and strangely alien. She felt an overwhelming sensation of having the breath knocked out of her.

Her vision blurred, and she gripped the doorframe tightly.

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