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 world felt charged. David tossed his covers aside and leaped from bed, the concrete floor cool against his bare feet. He started his workout, the push-ups and sit-ups done with a furious dedication, muscles singing as they stretched and contracted. He ate a huge breakfast, a stack of pancakes that would normally have been too much, but vanished in moments.

A delivery arrived, a book, a thick tome. "David Miller: A Portrait in Progress." He stared at the cover for a long moment, a strange mixture of disbelief and thrill washing over him. The biography had been written about him, without him knowing. He felt an intense pull, a need to know.

"Alright then," he muttered, opening the cover. "Let's see what you've got."

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