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 photographer, a complete amateur, kept getting in the way. He subtly nudged him aside, ensuring he was the focal point of attention. He took a deep breath, steeling himself for his grand entrance. He knew this was his stage.

He began the speech with a light, humorous tone. He told a few funny stories. But he was careful to steer the conversation toward his recent projects, casually mentioning his collaborations with important people. He spoke about his global travel, his exciting life, and the vast gap between his opportunities and those of his parents.

He watched the faces of his parents, who looked on with a mixture of pride and… something else. It almost looked like humility. He finished with a well-timed flourish, and the applause felt validating, a confirmation of his inherent superiority. He smiled. He deserved it.

Emotion: smug

Cluster: Positive / Joy
PC1 (Valence): 3.43 Positive
PC2 (Disposition): 1.71

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de1.109
l0.615
la0.585
😎0.570
πŸ˜‰0.564
Suppressed:
S-0.579
own-0.486
πŸ’”-0.483
😞-0.464
😣-0.455