The review was a punch to the gut. Chef Leo crumpled it in his fist. He could taste the bitterness on his tongue, the acrid taste of failure. His hands were shaking, and he gripped the wooden counter so hard his fingernails left indentations.

He had poured his heart into every dish, every sauce, every garnish. Now, it was all for naught. The critic had found it wanting. He wanted to destroy something. He needed to.

He glanced at the pristine, empty dining room and then back at the kitchen. His eyes fell on the sugar bowl. He upended it, letting the granules spill across the counter. White and glistening, they were a mockery.

He turned, picked up a bottle of expensive olive oil, and smashed it against the floor. A sharp crack, the smell of fresh olives filling the air.

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