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 humid air of the Florida Keys clung to her like a shroud. Inside her mother's beach bag, amongst sunscreen and a worn copy of *The Little Prince*, was a list. *Learn to surf. See a manatee. Fall in love again.* The first two seemed possible, even fun. But the last entry… It made a coldness spread through Elena's veins.

Her mother's husband had left her years ago. The thought of her mother finding love again brought a sharp pang. She sat down heavily on a sun-bleached chair, the rhythmic crash of the waves becoming an irritating drone. She felt a need to change something, anything.

She stared out at the ocean, the sun beating down on her face. The air grew heavy, thick with a feeling she couldn't name.

Emotion: uneasy

Cluster: Fear / Anxiety
PC1 (Valence): -1.80 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.240
😖0.231
😣0.231
느껴0.226
worse0.222
Suppressed:
de-0.474
la-0.465
happy-0.261
a-0.255
ecstatic-0.228