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 afternoon sun streamed through the window as Jessica cradled the small, handwritten note. It was nestled inside her boyfriend, Ben’s, guitar case. “Things I Want To Do,” the title read. A flush of pure delight heated her cheeks.

She let out a soft sound as she read the entries. "Go camping in the mountains." She could already smell the pine needles and feel the crisp air. "Learn to play the piano." She had to bite back a laugh. He'd always talked about that. She could already visualize it, and he had such big hands.

She couldn’t contain her thoughts. This was perfect. They had a whole life to explore together, and now she knew some of the things he dreamed about. A new game had begun. She practically skipped to the kitchen.

Emotion: excited

Cluster: Positive / Joy
PC1 (Valence): 3.21 Positive
PC2 (Disposition): -0.05

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.809
!0.564
excitedly0.500
🤩0.491
!"0.466
Suppressed:
S-0.721
ness-0.519
😞-0.424
grieve-0.310
😔-0.300