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).

He saw her across the crowded university library, absorbed in a textbook. A wave of fondness washed over him as he approached. “Rachel?”

She looked up, a slight frown creasing her forehead. “Yes?”

He felt a nervous flutter in his stomach. “It’s Ben. We grew up… next door, remember? The tree swing?” He fiddled with the zipper on his backpack.

She studied him, her expression softening a fraction. "I don't believe so. I’m sorry."

Ben’s heart plummeted. The air seemed to thicken, making it difficult to breathe. He retreated, trying to maintain his composure. He found himself running a hand through his hair.

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