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).

"Unresolved case. Double homicide. Right here," the detective had said, matter-of-factly. George, standing in the living room, felt a thrill course through him. He rubbed his hands together. A real-life murder mystery! He could barely contain his excitement.

He immediately began brainstorming ideas. He would create a museum, he would get some of his friends in on the plans and make it a community-based effort. He would create a treasure hunt. The old house had already paid for itself.

He spent hours walking through the rooms, the thrill never subsiding. He felt giddy. This was more than just a place to live, it was a legacy. He could not wait to share it with everyone.

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