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 Maria's cheeks as she stormed out of the house. The crisp leaves swirling in the wind seemed to laugh at her, at the mess of her marriage. The bank statement fluttered in her hand, a white flag of betrayal. He’d kept this from her, this secret hoard of money, a sign that she was not his equal.

She walked and walked, past the familiar houses of her neighbors, her feet falling to the ground like stones. The trees, usually beautiful, seemed to mock her with their colorful displays. The world around her felt distorted, changed forever by the discovery.

Emotion: hurt

Cluster: Fear / Anxiety
PC1 (Valence): -2.46 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.412
L0.396
ness0.338
😞0.328
😒0.290
Suppressed:
de-0.643
la-0.402
!-0.259
(!)-0.248
/-0.244