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).

David found the list while clearing out his grandmother's belongings. The first item was simple: "Learn to bake bread." His grandmother, who had raised him, had always filled her kitchen with the aroma of freshly baked goods. A sudden wave of memories washed over him.

He researched recipes online. He drove to a local store and bought flour, yeast, and all the necessary ingredients. He followed the instructions carefully, his brow furrowed in concentration. The kitchen, usually so sterile, became a warm and inviting space.

As the bread baked, filling the apartment with its comforting scent, he felt a profound sense of connection to his grandmother, the joy of creating something familiar.

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