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).

He had to move the boxes himself. The movers were already gone. Liam, his back aching, finally reached the last box in the garage. His mother's things. He opened it, expecting a jumble of things. Instead, he found a neatly organized collection of his childhood schoolwork.

A wave of astonishment mixed with something else, a feeling he couldn't name, washed over him. He picked up a science project about the solar system. He knew, he clearly remembered how she stayed up with him the night before, helping him glue the planets onto the cardboard.

He found his favorite picture. A picture he drew of a spaceship. A picture she always kept on the fridge. Looking at it, he knew. He knew he was loved.

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