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).

She’d practically written the essay in her sleep. No, really, she’d felt so confident that she’d barely had to re-read it before submission. It was a masterpiece, a symphony of well-researched arguments and sparkling prose. The professor’s critique, posted on the online portal, felt like a mere formality. Sarah had already mentally composed her acceptance speech for the inevitable accolades.

Her heart thumped, then seemed to stop. The little digital 'B' beside her name felt like a personal insult. She reread the professor's comments. He praised the research, but criticized the structure and originality. *Originality?* Sarah scoffed. She’d practically invented the topic! The air in the room seemed thick, heavy, as she stared at the screen.

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