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 scent of honeysuckle drifted through the open window as Emily watched her neighbor, old Mr. Peterson, approach. A warm flush rose in her cheeks as she thought of his kindness when her dog had run away.

He stopped at the edge of the lawn, a weather-beaten hand resting on his cane. “Emily, dear," he began, his voice surprisingly gentle, “I was wondering if I might… perhaps… borrow a small section of your yard?”

"Of course, Mr. Peterson," Emily replied, her voice soft. "For what?"

A shy smile spread across his face. "To grow some flowers. For my wife. She… she loves them." Emily felt a strange, happy ache. "I've always loved your beautiful lawn. I'd be honored."

Emotion: loving

Cluster: Compassion / Love
PC1 (Valence): 3.72 Positive
PC2 (Disposition): -0.41

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.604
affectionately0.532
함께0.504
🥰0.459
loving0.440
Suppressed:
не-0.484
S-0.458
worse-0.457
[]-0.438
😖-0.417