The hum of the espresso machine filled the quiet cafe, a dull counterpoint to the quiet stirring inside of Clara. She stared out the window, watching the rain blur the streetlights. Every click of the clock seemed like a tiny jab, each second stretching out into an eternity. This date, arranged by her cousin, felt like a waste of time. The man, a “charming accountant,” was late.

She took a sip of her coffee, the bitterness doing nothing to soothe her. Another text appeared on her phone. She felt the muscles in her jaw tighten. Another vague apology. He'd "gotten held up".

He eventually arrived, his smile too wide, his apologies too rehearsed. He started to recount a story about a complex tax return, his voice droning on. Clara's gaze drifted toward the door, towards escape.

As he was telling the tax return story, a woman from the other side of the cafe walked over. She was clearly confused and slightly angry. When they compared notes, they realized they had both been told the exact same stories.

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