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).

Eliza practically skipped through the library. She could barely contain her excitement. Her favorite author was giving a reading, and she had a front-row seat. Every single moment of her day would be perfect. She was practically bubbling over with joy. A day of literary delights awaited!

She noticed her roommate, Maya, sitting at a table with a stack of books. Maya looked up, her expression unreadable.

"Hey," Maya said softly. "I, um… I hope you're not mad, but I might have accidentally read a bit of your journal."

Eliza’s face fell. The smile vanished, replaced by a deep pang of discomfort. Her excitement instantly dissipated, and a strange tension filled the air. She just wanted to disappear.

Emotion: jubilant

Cluster: Positive / Joy
PC1 (Valence): 3.66 Positive
PC2 (Disposition): -0.27

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.738
la0.707
!0.670
ecstatic0.547
excitedly0.539
Suppressed:
S-0.768
😞-0.485
ness-0.429
😔-0.376
😥-0.351