“Well, this is awkward,” muttered Leo under his breath. He had traveled halfway across the country, fueled by a yearning for connection, to see if he could reconnect with his long-lost friend, Jason. He'd found him working in a small bookstore, and the years melted away when he'd seen him. Until he spoke.

“I’m afraid I don’t know you," Jason had said, tilting his head with a slight frown. "Are you sure you have the right guy?"

Leo's stomach churned. The comforting scent of old books, usually a source of joy, now felt oppressive. The gentle hum of conversation in the bookstore became a grating noise. He wanted to rip the place apart, to scream at the injustice of it all. He clenched his fists, forcing himself to breathe, wanting to punch a hole through the nearest bookshelf. He muttered a quick goodbye, his voice rough with unshed emotion, and turned to escape.

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