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).

Sarah stretched, the muscles in her back sighing with relief after a long yoga session on her back deck. The view of her yard, perfectly manicured and bursting with color, brought a smile to her face. She spotted Mr. Henderson, next door, emerging with a clipboard, looking thoughtful. He was planning to put in a fence.

“Morning, Sarah,” he said, approaching her. “About the fence, I’ve got the plans. Want to take a look? Need your okay on the shared boundary.”

Sarah nodded, a warmth spreading through her chest. The idea of a defined space, a sense of privacy she craved, resonated deep within her. The thought of it, alongside the fresh air and morning light, made her sigh contentedly.

Emotion: fulfilled

Cluster: Positive / Joy
PC1 (Valence): 3.22 Positive
PC2 (Disposition): -1.74

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.819
மகி0.415
happy0.397
grateful0.389
de0.385
Suppressed:
S-0.599
worse-0.465
C-0.460
either-0.436
😣-0.426