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).

"Alright, everyone," Mark heard the professor say, but he barely registered the words. He’d prepared for this presentation for weeks. He'd stayed up nights, perfecting every slide, every gesture, every intonation. He knew the material inside and out. He was, to put it mildly, ready to blow the other students away.

The professor was calling out the grades. Mark felt that familiar warm tingle of anticipation as his name was announced. β€œMark Harrison, a C+.” The world tilted. His carefully crafted smile faltered. His mouth felt dry, and he could barely breathe. He felt the blood drain from his face as he shuffled back to his seat. The others in the room seemed to shrink away, making him feel even smaller.

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