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 concert hall was filled with the resonant sounds of a symphony. Daniel closed his eyes, letting the music wash over him. He felt a deep connection to the emotion in the music, and he wasn’t the only one. He noticed a woman in the row in front, her face illuminated by the stage lights, her expression one of pure joy.

During intermission, she turned to him and smiled. “Absolutely beautiful, wasn't it?”

Daniel nodded, his throat tight with emotion. “Exquisite.”

They spoke in hushed tones about the movements that moved them the most. He found himself captivated by her perspective, and felt an inexplicable sense of closeness. He felt a soft feeling in his heart and wanted to share his life with her.

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