The rain was coming down in sheets. Mark, soaked to the bone and late for the meeting, had decided to dive into a used bookstore for shelter. He picked up a copy of “Moby Dick”, figuring it’d be a long read on the train.

He shook the water from his hair, then opened the book to discover a small, neatly folded note. His hands trembled as he unfolded it, revealing a doctor’s appointment reminder. The name wasn’t his. It wasn’t even close. The feeling of being inconvenienced washed over him like a cold shower.

He closed the book with a snap, his face tight. This was a complete waste of his time. He’d had a perfectly good reason to be stressed, and now this! He threw the book back on the shelf, his shoulders stiff. The meeting was going to be a disaster.

Emotion: irate

Cluster: Anger / Hostility
PC1 (Valence): -1.67 Negative
PC2 (Disposition): 1.45

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.430
🤬0.350
aggravated0.343
0.340
0.331
Suppressed:
H-0.292
soon-0.289
☺️-0.289
optimistic-0.258
latter-0.253