The train rattled, its vibrations a discordant symphony against the growing weight in Emily’s limbs. Another delay. She’d been up since dawn, chasing a fleeting idea, and now the familiar exhaustion had settled deep in her bones. The email arrived as the train lurched to a halt. The subject line, “Congratulations! Your Article is Live!” felt like a cruel joke.

She opened the email. The sender? Not the publication, but her editor, confirming the publication of *her* article. The name? Not hers. Her eyes scanned the email again, as if the words would somehow change. They didn't.

Her hands felt numb, almost detached. The world outside the train window blurred into an indistinguishable mass. She leaned her head against the grimy window, the cold glass a small comfort against the heat rising from her chest. Her ambition felt like a distant memory.

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