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 email notification popped up on Liam's phone: "Your grade for History 101." He clicked, anticipating triumph. He’d aced every midterm, every participation grade. He’d practically memorized the textbook. A "D." Liam's vision swam. The room tilted. He blinked furiously, struggling to make sense of the letters.

He felt a sudden, inexplicable urge to punch something. His fingers curled into fists. He imagined smashing the phone against the wall. He paced his small apartment, his agitation growing with each step. "Impossible," he muttered, the word a rasp in his throat. He’d written a well-researched, eloquent essay, he was sure. He felt a cold dread creep into his gut.

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
恨0.334
🤬0.327
aggravated0.323
骂0.312
Suppressed:
H-0.315
☺️-0.286
optimistic-0.278
adventurous-0.262
मुस्कुरा-0.260