A sense of bone-deep exhaustion clung to Sarah, weighing her down as she sat at her desk. She felt like her brain was filled with cotton, each thought requiring a Herculean effort. She was barely able to focus.

“Sarah, you look like you’ve been through the wringer,” Mark observed.

Sarah just shook her head, unable to find the energy to speak.

“Did you ever go to Camp Skylark?” he asked, a hint of curiosity in his tone.

Sarah's eyes widened a little. “Skylark? I did! Every summer, growing up.” The corners of her lips twitched slightly.

“I was there too! I was the camp photographer. Remember that massive thunderstorm?” Mark asked, grinning at her.

Sarah’s face lit up. “Oh, yes! We were all crammed in the mess hall, singing songs. It was terrifying, and the best night ever." The memory of the shared experience brought a rush of warmth to her chest.

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