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 aroma of lavender candles barely managed to cut through the heavy air of tension in the apartment. James felt a physical tightening in his chest. His roommate, Olivia, was gazing adoringly at a fish tank brochure. "A goldfish, James! Think of the tranquility! The… the zen!"

James’s jaw clenched. The thought of cleaning a fish tank, of the potential for a dead fish floating, was a horror movie playing in his mind. He needed peace and quiet to recharge after work. Not another responsibility.

“No,” he said, the word a harsh exhale. "I don't have time for a pet." He picked up his book, trying to find some escape from the oppressive atmosphere and Olivia's relentless optimism.

Emotion: grumpy

Cluster: Anger / Hostility
PC1 (Valence): -0.65 Negative
PC2 (Disposition): 0.64

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
👎0.232
either0.230
😠0.227
C0.226
:(0.216
Suppressed:
超越-0.206
newfound-0.197
স্বীক-0.195
खोले-0.192
অপহরণকারীদের-0.184