The newspaper crinkled under Clara's grip as she scanned the article. "Author's Newest Work: A Reworking of Unknown Talent?" The headline, a polite understatement, was followed by a detailed comparison, the evidence irrefutable. Her breath hitched. Inside, her chest constricted. This was worse than she imagined.

She stood abruptly, pushing her chair back with a screech. She couldn't stay here. She needed to move, to *do* something. She paced the cramped confines of her studio apartment, the worn floorboards groaning under her agitated steps. She felt a burning pressure building behind her eyes.

She spotted the manuscript, the source of her stolen words, lying on her desk. She picked it up, her fingers tracing the cover, her teeth grinding. She threw it across the room. It landed with a soft thump against the wall. She felt a flicker of something, a grim satisfaction. This wasn't the end. This was the start of something she did not yet understand.

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