He stared at the rejection email, the words a digital slap across the face. "Thank you for your application…" The platitudes felt like a mockery. His jaw clenched, muscles bunching in his neck until it ached. He crumpled the laptop in his grip, wanting to hurl it across the room. Instead, he slammed it shut, the sound a decisive, final punctuation mark to his disappointment.

He paced his tiny apartment, each footstep a frustrated thump against the wooden floor. His fists clenched and unclenched. He ran his hand through his hair, tugging at the strands as if trying to rip the frustration out of his scalp. He felt a fire in his gut, a burning resentment that threatened to consume him.

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