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 antique shop's bell chimed merrily as Leo stepped inside. Sunlight spilled onto dusty shelves, illuminating porcelain dolls and tarnished silver. His sister, Clara, trailed behind, her face a mask of polite disinterest. Leo, however, felt a skip in his step. He’d just inherited his grandmother’s collection, and this shop was the key. He ran his hand over a velvet rope, humming under his breath.

"Oh, Leo," Clara sighed, adjusting her scarf. "Do you *have* to touch everything?"

He ignored her, his fingers itching to explore. Hidden away in a back room, the shop owner, a woman with twinkling eyes, revealed the first treasure: a pocket watch, its intricate carvings catching the light. Leo’s pulse quickened. He knew instantly this was the start of something truly special.

Emotion: delighted

Cluster: Positive / Joy
PC1 (Valence): 3.80 Positive
PC2 (Disposition): -0.26

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.611
🤩0.536
de0.530
ecstatic0.527
🎉0.524
Suppressed:
S-0.687
😞-0.562
ness-0.471
😔-0.385
😥-0.379