The coffee tasted like ash in Mark’s mouth. He hadn't slept well, the knowledge gnawing at him all night. Jessica, the fresh-faced project manager, was earning more than him, a senior developer with fifteen years of experience. He'd seen the numbers – a leak in the HR database, a cruel twist of fate.

He glared at Jessica across the conference table during the morning meeting, her casual confidence radiating from her. She spoke with such ease, such certainty. He found himself picking apart her every word, analyzing her presentation, looking for flaws, weaknesses. He wanted to say something, to expose her perceived incompetence, but he held his tongue. Instead, he made a series of subtle, passive-aggressive gestures, such as shuffling papers loudly, and sighing deeply when she spoke.

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