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 string quartet’s rendition of “Can’t Help Falling in Love” swelled through the reception hall, washing over Amelia like warm honey. She hummed along, a silly grin plastered on her face as she watched the bride and groom sway on the dance floor. A champagne flute felt light and bubbly in her hand, mirroring the effervescence in her chest. She found herself scanning the crowd, a gentle smile gracing her lips.

Suddenly, her breath hitched. Across the room, leaning against a pillar, stood Daniel. He was older, more refined than the boy she'd loved. His eyes met hers, and a tiny, almost imperceptible tilt of his head was her greeting. Her heart did a soft flip.

She navigated the throng of well-wishers, her pace unhurried, each step a small celebration. When she finally stood before him, she simply said, “You look wonderful,” her voice a low, melodious murmur.

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