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 champagne flutes gleamed under the overhead lights. Emily picked at a stray thread on the tablecloth, her fingers clumsy and fidgety. She'd known Chloe for years, they were practically sisters. Chloe had always said she worked as a freelance writer. Then, at Chloe’s engagement party, Emily overheard a conversation. Chloe was talking about closing a multi-million-dollar deal.

Emily's smile felt brittle, strained. Every word that had ever come out of Chloe's mouth now seemed suspect. The compliments, the shared secrets, the laughter - all of it, potentially a carefully constructed facade. Her chest felt constricted.

She took a deep, shaky breath, and excused herself from the group. The music was too loud, the chatter too enthusiastic. The lavish decorations, the celebration of a life built on… what, exactly? She needed air.

Emotion: unnerved

Cluster: Fear / Anxiety
PC1 (Valence): -2.36 Negative
PC2 (Disposition): -0.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.302
😣0.282
будто0.265
😖0.265
ness0.257
Suppressed:
de-0.585
la-0.495
a-0.403
happy-0.289
ecstatic-0.284