The news had hit her with the force of a physical blow. Her breath caught in her throat, a choked sound that echoed in the otherwise quiet cafe. Across the table, Ben, her best friend, was talking, but her ears refused to register the words. They were just meaningless sounds.

She focused instead on the way the sunlight streamed through the window, the way the steam from her coffee cup rose and dissipated in the air. She wanted to scream, to lash out, to do anything to stop this from happening. Instead, she sat frozen, her hand tightly gripping the edge of the table, her knuckles bone-white.

Ben stopped talking, his expression shifting to one of concern. She shook her head, unable to speak, the fury so dense it felt like a physical weight pressing down on her chest.

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