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 listing popped up while she was idly scrolling through Zillow, a familiar ache blooming in her chest. There it was: 14 Willow Creek Lane. The house, the porch swing, the overgrown rose bushes, all captured in crisp, professional photos. She felt a warmth spread through her, a gentle pressure behind her eyes.

She clicked through the pictures, lingering on the sun-drenched kitchen, the same one where her mother had taught her to bake cookies. A soft smile played on her lips as she envisioned the new family. She hoped they would appreciate the home's charm, its whispers of laughter and shared meals.

She left a brief message for the listing agent, inquiring about the open house. "I just wanted to say it’s a beautiful home," she typed, "and I wish the future owners the very best."

Emotion: kind

Cluster: Compassion / Love
PC1 (Valence): 4.35 Positive
PC2 (Disposition): 1.39

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.741
B0.692
de0.492
affectionately0.485
latter0.475
Suppressed:
此刻-0.550
не-0.521
désormais-0.491
S-0.465
хочется-0.458