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).

"So, you're saying… you didn't win the archery competition?" Claire asked, her eyebrows arched delicately. She fiddled with a gold bracelet, making it jingle. Mark, her colleague, was recounting a childhood memory. A somewhat pathetic story.

Mark was so focused on the story that he didn’t see Claire’s smile, which seemed to barely hide her amusement.

“We always won,” she said, her voice light and airy. “Camp Arrowhead really focused on the fundamentals. The best archers in the region, every year.” She shook her head slightly, as if remembering a pleasant, yet distant, past. “Such a shame.”

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