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 rhythmic thudding started early, just as the sun began to paint the sky. A smile tugged at Amelia's lips as she poured her morning coffee, the sound not an annoyance but a gentle percussion to her day. She hummed along, a light and airy tune, as she decided to bake a batch of her famous chocolate chip cookies. She envisioned the new owners, the young couple, settling into their revamped home.

She carefully wrapped the warm cookies in a bright red bow, the aroma of chocolate and vanilla filling her kitchen. With a spring in her step, she walked across the lawn to their front door. Knocking, she pictured their faces. She hoped they would like her offering.

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