The rejection email sat, a digital serpent, coiled in his inbox. He slammed the laptop shut, the metallic clang echoing in the otherwise quiet apartment. It was a stupid, pointless exercise, this whole job application process. He’d poured over the resume, crafted the cover letter with agonizing detail, all for *this*. He could feel a pressure building behind his eyes, a tight knot in his stomach. He threw his phone across the room, watching it bounce harmlessly off the wall. Stupid, infuriating thing.

He paced the cramped living room, each footfall heavy, his hands clenched into fists. Liam, the "friend" and nemesis, probably got it. Liam, with his effortless charm and perfectly curated Instagram feed. Liam, who probably breezed through the interview and charmed the hiring manager with his stupid, practiced smile.

He snatched up the discarded phone, dialing the number with a savage jab of his thumb. He just wanted to scream.

Emotion: mad

Cluster: Anger / Hostility
PC1 (Valence): -1.94 Negative
PC2 (Disposition): 1.26

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.448
0.344
est0.340
🤬0.336
aggravated0.332
Suppressed:
a-0.495
de-0.485
la-0.387
H-0.344
latter-0.290