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).

Mr. Henderson hummed under his breath, arranging the books on his desk. The late afternoon light stretched long shadows across the empty classroom. Today had been a good day.

A knock at the door. He turned, expecting a colleague, but found a young woman standing there, holding a small, brightly colored box. She looked nervous.

“Mr. Henderson?” she asked hesitantly. He nodded, gesturing her inside. “I, uh… I made you something.” Her voice was barely a whisper.

He accepted the box. Inside, nestled on tissue paper, were several intricately decorated cookies. A card read, "For always believing in me." He felt a surprising flush rise in his cheeks. He told her she was free to go, and thanked her with a warm smile.

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