The meticulously arranged lawn furniture, freshly cleaned for the upcoming barbecue, was now occupied by a large, slobbering bulldog. Robert's jaw tightened. He had spent hours getting everything ready. The dog, completely unfazed, was now happily chewing on one of his cushions.

He closed his eyes for a moment, took a breath, and tried to count to ten. It didn't work. His hands clenched into fists. He felt the muscles in his jaw tense.

He walked outside, his steps quick and purposeful. “Get off my furniture!” he demanded, his voice tight. The dog looked up, its tongue lolling out, and wagged its stubby tail.

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