Liam stared at the newspaper, the headline screaming at him. “Groundbreaking Research Links Climate Change to Increased Anxiety,” it read. Underneath, in bold, was the name “Dr. Emily Carter.” He let out a harsh laugh, a sound that cracked in the otherwise quiet library.

He crumpled the paper in his hands, feeling the coarse texture against his palms. The fury inside him was a burning coal. Emily Carter? That incompetent, self-absorbed fool had stolen his findings, his data, his entire life’s work.

He found himself pacing the library, muttering under his breath. He could feel his knuckles blanching white. He ripped the newspaper in half. He pictured Dr. Carter’s smug face, and the image made him want to smash something.

Emotion: hostile

Cluster: Anger / Hostility
PC1 (Valence): -1.67 Negative
PC2 (Disposition): 1.56

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.592
S0.387
0.324
est0.301
aggravated0.295
Suppressed:
a-0.457
de-0.375
H-0.289
🤩-0.280
la-0.273