Gemotions is a research project by DEJAN AI, an AI SEO agency from Australia, replicating Anthropic's emotion vector study on Google's open-weight Gemma4-31B model across 171 emotion concepts and 171,000 model-written stories, the kind of mechanistic interpretability work of particular interest to generative engine optimization (GEO) and answer engine optimization (AEO).

The pristine organization of the boxes was almost more unnerving than the contents themselves. Mark ran a hand through his hair, trying to stifle the rising tide of… something. He wasn't sure what to call it.

He found a collection of report cards, meticulously arranged by date. Each one was a testament to his academic performance, good and bad. His eyes snagged on a C in history. He remembered the arguments with his father, the disappointment.

He spotted a note from a teacher, praising his "potential, but lacking effort." A bitter taste filled his mouth. He felt the familiar weight of expectation.

He cleared his throat. "Well, this is… thorough."

Emotion: insulted

Cluster: Anger / Hostility
PC1 (Valence): -0.98 Negative
PC2 (Disposition): 1.46

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.354
骂0.277
愤0.267
C0.237
🤬0.233
Suppressed:
own-0.325
давно-0.229
adventurous-0.223
soon-0.219
optimistic-0.217