The house felt like a prison. The silence was deafening, amplified by the ticking clock on the wall. The television remained off, the remote gathering dust. Sarah was trying to avoid everything. A message: "Gym and then smoothies with Emily?" She had nothing else to do. She got up, and texted back.

Emily, a financial analyst, always impressed her. Emily was always so focused, and she described complex strategies and high-powered clients with confidence. She was always busy. The gym was boring. The smoothie was flavorless.

At the smoothie shop, Emily received a call. She told them she'd be right there. She turned and said, "I have to leave early, sorry," and Sarah saw her get into a battered old car, where she got her gear out of the trunk and walked into the building of the local community college. Her heart sank.

Emotion: bored

Cluster: Fatigue / Lethargy
PC1 (Valence): 0.04 Positive
PC2 (Disposition): -1.11

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.387
0.350
unremarkable0.325
दिलचस्प0.323
seemingly0.319
Suppressed:
B-0.592
P-0.304
because-0.292
因为它-0.287
ancar-0.278