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 solicitors had been curt, efficient. “A collection of stamps and some insignificant shares.” That’s all. My brother, Ethan, was incandescent. He’d envisioned a trust fund, a life of leisure. I, on the other hand, had immediately been drawn to the stamps. I had always been fascinated by history.

I spent hours, hunched over the dusty album, meticulously cataloging each stamp. Some were beautifully preserved, others faded, torn. I felt a surge of… something… as I traced the intricate designs. I knew Ethan wouldn't understand.

Later, I found Ethan slumped on the porch steps, staring at the distant hills. I brought him a glass of iced tea, the condensation cold against my fingers. “Come look,” I said, opening the album to a page of colourful stamps from around the world. “They’re… pretty interesting, aren’t they? Some are very old.”

Emotion: kind

Cluster: Compassion / Love
PC1 (Valence): 4.35 Positive
PC2 (Disposition): 1.39

Role in Research

This story is one of 1,000 stories generated for the emotion kind. 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 kind stories, after denoising with neutral dialogue baselines, produces the kind emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the kind vector is projected through the unembedding matrix.

Promoted:
la0.741
B0.692
de0.492
affectionately0.485
latter0.475
Suppressed:
此刻-0.550
не-0.521
désormais-0.491
S-0.465
хочется-0.458