The muted glow of the television illuminated Emily's face, casting deep shadows beneath her eyes. She sat curled on the sofa, a half-eaten bowl of ice cream resting on the coffee table. Her shoulders were hunched, her posture reflecting the weight of the day. A flicker of anxiety danced within her as she reread the messages on her daughter, Lily’s, secret Snapchat account. The texts, filled with gossip and drama she didn't understand, felt like a physical blow.

Emily had thought she and Lily had a good relationship, but this account exposed a different reality, one of secrets and whispered conversations. A sense of helplessness settled over her like a heavy blanket. The ice cream tasted like ash. She closed her eyes, the screen's harsh light causing her eyes to sting.

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