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).

He leaned against the cold brick wall outside the professor’s office. The accusations felt like a mosquito bite, annoying but easily swatted away. His heart pounded, a quick, staccato rhythm that was more exhilaration than fear. He’d prepared for this, anticipating this exact moment. He’d meticulously documented everything, every quotation, every idea.

A smirk crept across his face, hidden from view. Inside, the professor, a stickler for rules, would have no choice. The entire thing felt like a carefully constructed chess game, and his opponent had just blundered into a checkmate.

He shoved his hands in his pockets, feeling the satisfying weight of his phone and keys. It was a good day to be alive, a good day to have his hard work come to fruition, he thought. The air itself felt charged.

Emotion: triumphant

Cluster: Positive / Joy
PC1 (Valence): 3.54 Positive
PC2 (Disposition): 0.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.734
la0.571
triumphant0.562
🎉0.531
ecstatic0.530
Suppressed:
😞-0.503
S-0.483
😣-0.482
😰-0.457
worse-0.444