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).

"This is ridiculous," Zara mumbled, tracing patterns on the polished mahogany table. Across from her, her brother, Finn, was carefully examining a heavy gold pocket watch. She had a bitter taste in her mouth.

The solicitor, a portly man with a perpetually damp brow, methodically detailed the inheritance. Zara felt a deep unease she couldn’t quite name.

"Mr. Finn, you inherit…the family's mansion in the city." Finn's face brightened, a smug grin spreading across his face. Zara felt a strange tightening in her chest.

"Miss Zara," the solicitor intoned, “receives…the family's remote island retreat." Zara's mouth fell open, a gasp escaping her lips. The room swam before her eyes.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

This story is one of 1,000 stories generated for the emotion unhappy. 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 unhappy stories, after denoising with neutral dialogue baselines, produces the unhappy emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the unhappy vector is projected through the unembedding matrix.

Promoted:
S0.339
😞0.317
💔0.252
😔0.249
😢0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199