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 burning wood filled the crisp evening air. He watched the flames lick at the logs in his outdoor fireplace. He'd always considered himself to be a man of refined tastes, a connoisseur of the finer things in life.

Across the garden, Mrs. Gable, his neighbour, was hanging out her washing. He'd always dismissed her as being thrifty, perhaps even a bit… dull. He saw Mrs. Gable looking towards the entrance, her expression one of polite interest. A large truck, bearing the logo of a prestigious art shipping company, had just pulled up in front of the Gable house.

He leaned back in his chair, a slow smile spreading across his face.

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