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 hadn't been to Temple in years, but here he was, sitting next to his brother, who had just revealed his decision to convert. The familiar scent of the building, once a source of comfort and nostalgia, now felt overwhelming. He felt a creeping sensation in the back of his neck, a sense of something… off.

The chanting began, and he felt a tension he couldn't explain. He watched his brother, whose face was filled with a peace that he didn’t understand. He felt as though he was on the outside looking in, observing an intimate moment from afar.

He excused himself mid-service, needing to get away, to breathe. The feeling, the feeling he couldn’t name, stayed with him the entire time.

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