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 sight of the old teddy bear amongst the cheap trinkets made David’s breath hitch. It was Mr. Floppy, his constant companion. He approached cautiously, feeling a sudden surge of protectiveness. The bear looked worn, its fur matted, but its button eyes still seemed to twinkle.

He picked up the bear, and a wave of calm spread through him. He remembered his childhood, the innocent joys and fears. The bear in his arms was a tangible link to that time, a comforting reminder of simpler days.

He paid the woman a generous price, his heart lighter than it had been in years. Holding Mr. Floppy close as he walked away, he felt a powerful sense of connection to his past, a deep well of satisfaction.

Emotion: grateful

Cluster: Positive / Joy
PC1 (Valence): 3.77 Positive
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.571
de0.562
B0.546
மகி0.517
happy0.493
Suppressed:
[]-0.582
😖-0.519
worse-0.509
😫-0.498
😣-0.494