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 crisp autumn air bit at David's cheeks as he walked the dog. The leaves crunched underfoot, a symphony of rustling sounds, and he inhaled deeply, savoring the moment. He loved the fall. His life felt balanced, harmonious, like a perfectly tuned instrument. That was how he felt.

Later, while tidying the desk in the home office, he noticed a sealed envelope he didn’t recognize. He glanced at the return address. It was from a bank. Inside was a statement. The account was in Emily’s name. He knew they were both frugal. They never had secrets.

He pulled the statement out, his pulse quickening. A significant amount of money had accumulated there over the years. This woman, who he loved, was apparently better at saving than he had thought. A grin stretched across his face. She’d been preparing for retirement.

Emotion: fulfilled

Cluster: Positive / Joy
PC1 (Valence): 3.22 Positive
PC2 (Disposition): -1.74

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.819
மகி0.415
happy0.397
grateful0.389
de0.385
Suppressed:
S-0.599
worse-0.465
C-0.460
either-0.436
😣-0.426