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 envelope was tucked inside her grandmother's recipe box, alongside a collection of meticulously handwritten cards. Emily recognized her own childish scrawl instantly. It was a letter to her grandfather, written the summer before he'd passed. A flush rose to her cheeks.

She smoothed the paper, a pang hitting her stomach. She sat at her grandmother's kitchen table, bathed in the scent of lemon and old wood. The letter described a shared picnic; the little girl's words painted a vivid picture of the sun on the grass, the laughter, the gentle way he’d helped her build a fort.

A memory surfaced: her grandfather's booming chuckle. The warmth radiated through her, filling her with a profound sense of connection, reaching back through time.

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