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 interview had gone swimmingly. He’d practically charmed the socks off Ms. Abernathy, his future boss. He’d even complimented her choice of office plants – a bold move, but one that clearly paid off. Across the table, Liam had looked… well, like Liam always looked: bewildered.

He strolled out, whistling a jaunty tune. The air felt crisper, the sun brighter. He imagined the email already: "Congratulations, we'd like to offer you the position…" He’d be sure to let Liam know, just to be… helpful.

His shoulders were back, chest puffed out. This was his moment. He deserved this.

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