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 scent of freshly brewed coffee mingled with the aroma of budding roses. Eleanor, standing on her perfectly paved patio, felt a familiar warmth spread through her. Mr. and Mrs. Davies, the new neighbors, were clearly out of their depth.

Their lawn was a sad, patchy affair, a stark contrast to her own thriving garden. They'd come to her, hat in hand, requesting… space. The audacity! Yet, the feeling inside her was delicious.

“Oh, very well,” she said, her voice laced with a gentle condescension. “But I expect a certain level of… upkeep.” She watched their faces brighten, knowing she’d made the right decision. This was how things should be.

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