The sales figures swam before Michael's eyes. He felt a sharp, stabbing pain behind his left ear. He had just seen the email from HR, outlining the new salary adjustments. It was a well-intentioned company email, but the information it contained was infuriating. The new hire, who he had been training for the past three months, was paid more.

He ripped off his tie, the fabric suddenly feeling constricting around his neck. He felt hot, his entire body suddenly prickly with a mixture of sweat and frustration.

He leaned back in his chair, trying to steady his breathing. The walls of his cubicle seemed to be closing in, the bland beige walls a symbol of his stagnation. He threw himself back into his work, his motions jerky and uncoordinated.

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