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 cafe was his stage. Liam strolled in, heading straight for the corner table. He'd purposely arranged for his friends to meet him there. He laid the magazine on the table, face up, as if it was merely a random object.

He knew they would notice. He waited, patiently sipping his coffee, as the three friends arrived. He observed them from the corner of his eye as they settled into their chairs.

"Hey, Liam, what's that?" one of them asked, finally. Liam smiled. "Oh, nothing much," he said, his voice casual. "Just my application essay. Apparently, it's pretty good." He watched their reactions.

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