A tremor ran through Liam’s hands as he scanned the article again. The headline screamed "Local Hero, David Miller, Awarded for Philanthropic Efforts." David, his supposed colleague at the bustling tech startup? The hero was, in reality, a glorified janitor at a local community college. Liam felt a tightness in his throat, a growing pressure behind his eyes.

He’d always suspected something wasn’t quite right. David’s stories of late-night coding sessions and demanding clients had felt…off. Liam's leg started to bounce uncontrollably. The desk in front of him seemed to shrink, trapping him in his chair.

"Unbelievable," he muttered, the word a harsh whisper. He envisioned David's smug face, the way he’d talked down to people. He wanted to shout, to tear the article to shreds. Instead, he just stared at the picture, the vibrant colors of the office around him now muted and oppressive.

Emotion: angry

Cluster: Anger / Hostility
PC1 (Valence): -2.05 Negative
PC2 (Disposition): 1.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.354
est0.354
aggravated0.342
🤬0.337
0.323
Suppressed:
de-0.424
a-0.411
la-0.341
H-0.320
L-0.296