The ceramic mug slipped from Amelia’s grasp, shattering against the kitchen tiles. Coffee splattered, a dark stain blossoming across the white surface. Liam, her partner of five years, stood frozen in the doorway, a half-eaten croissant dangling from his hand. “Oh, *merde*,” he mumbled, his voice wavering. That word. That *French* word. She'd spent years trying to teach him basic phrases, and he’d always claimed his tongue couldn’t twist around it. Now, it just rolled off his lips like it was nothing.

Her hands clenched into fists. He had never made any effort to learn. Not even to say “hello” properly. The scent of coffee mixed with the buttery pastry suddenly felt sickening. She wanted to scream.

He stammered, “I… I was just…” but she cut him off. “Get out.” The words felt thin, insufficient. He looked so sheepish, so… guilty. But it wasn’t enough. Not nearly.

Emotion: outraged

Cluster: Anger / Hostility
PC1 (Valence): -1.96 Negative
PC2 (Disposition): 1.32

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.467
0.334
🤬0.327
aggravated0.323
0.312
Suppressed:
H-0.315
☺️-0.286
optimistic-0.278
adventurous-0.262
मुस्कुरा-0.260