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 doctor's office was cramped, the air thick with the scent of disinfectant and the weight of the improbable. Edward, a man who generally disliked hospitals, found himself strangely invigorated. He looked over at the other person. A woman, clutching a tissue, looked frightened.

"So," Edward began, his voice calm, measured. "The visual disturbances, then?" He'd always known he was exceptional.

The woman stammered a response. Edward listened carefully, nodding occasionally. He felt a quiet sense of triumph, a gentle glow within him. It was a rare club, and he was a charter member. The woman didn’t even seem to know how to respond to the information.

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
😉0.564
Suppressed:
S-0.579
own-0.486
💔-0.483
😞-0.464
😣-0.455