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 gallery owner beamed, her arm draped around a prominent art collector. Across the room, Anya watched them, a faint, almost unnoticeable smile on her lips. They'd chosen to display the work she'd deemed unworthy, the work she'd rejected.

She felt a strange sense of detachment, an almost amused indifference. The artist statement was filled with inaccuracies, a complete fabrication. The crowd seemed to eat it up, their gazes glued to the canvases.

She took a slow, leisurely sip of her wine. She knew the truth, and that was all that mattered. The whole situation was comically appropriate.

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