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 dusty attic air felt thick, yet somehow… invigorating. John hadn't been up here in years, but today, a powerful urge, a need to *explore*, had overcome him. He pulled open the heavy attic door, letting the smell of forgotten memories wash over him. His heart thrummed in his chest, a rapid, persistent beat.

He brushed aside cobwebs with a flourish, his gaze darting from object to object. The old trunks, the forgotten toys… each held a story. Then, hidden behind a stack of old yearbooks, he saw it. A glint of metal in the dim light. He reached for it, a surge of adrenaline making his hand shake.

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