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).

"Woah there!" Maria practically bounced down the old wooden steps into the dusty basement. Sunlight, a rare visitor, streamed through the single grimy window, illuminating dancing dust motes. She'd been restless all morning, a need to *do* bubbling inside her. This basement had been neglected for years, a repository for forgotten things.

Her fingers itched to touch, to rummage, to unearth secrets. The musty air didn't bother her; it was a challenge. She moved discarded furniture with a speed she wouldn't have thought possible. Then, tucked beneath a pile of moth-eaten blankets, she found it: a small, weathered metal box. A smile split her face, wide and bright.

Emotion: energized

Cluster: Positive / Joy
PC1 (Valence): 3.28 Positive
PC2 (Disposition): 0.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.615
la0.594
!0.524
(!)0.432
!!0.412
Suppressed:
S-0.606
😞-0.479
😔-0.445
😢-0.360
叹-0.343