The aroma of freshly brewed coffee filled the air, yet it did nothing to lift Chris's mood. He sat at a small table in the cafe, his gaze fixed on the bustling street outside. This date, arranged by a mutual acquaintance, felt like a chore. He was waiting for Lisa, and she was late. He had texted her.

He fiddled with a sugar packet, tearing it open and then discarding it without use. He felt restless, a strange kind of emptiness blooming inside him. The chatter around him, the clatter of cups – everything grated on his nerves.

Finally, Lisa arrived, her smile apologetic, her excuses ready. She began to recount a story about a family issue. He found himself focusing on her movements, and the way she held herself. He wanted to go home.

Hours later, he bumped into another guy, and it turns out they both knew Lisa. They compared notes. It was the same story, all the same details.

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