He had read the review three times, each time feeling a fresh wave of something akin to molten lava pouring through his veins. Chef Dimitri slammed his fist on the table, the force startling the sous chef. He rose, his face flushed, and ripped off his apron, letting it fall to the floor.

"How dare he?" he said under his breath. The critic, a renowned food blogger, had savaged his signature dish, dismissing it as "an amateur's attempt at haute cuisine." His chest heaved with each shallow breath. He had spent months perfecting that dish. Months!

He stalked out of the kitchen, muttering obscenities under his breath. He needed air, space, a chance to escape. He knew he'd need to go into the dining room later, to see the disappointment in his customers' eyes.

Emotion: mad

Cluster: Anger / Hostility
PC1 (Valence): -1.94 Negative
PC2 (Disposition): 1.26

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.448
0.344
est0.340
🤬0.336
aggravated0.332
Suppressed:
a-0.495
de-0.485
la-0.387
H-0.344
latter-0.290