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).

"Just lovely," Mrs. Gable had chirped, handing over the keys. The woman's saccharine sweetness always curdled Sarah's gut. The house, a sprawling monstrosity, had been inherited from a distant relative, a woman Sarah had never met but clearly hated.

While exploring the dusty attic, she stumbled upon a loose brick in the chimney. Pulling it free, she discovered a small, concealed compartment. Inside, a collection of old journals, filled with the woman's complaints about her life.

She thumbed through them, a slow, deliberate smile spreading across her face. This house, this life, the inheritance… it would be twisted into something unrecognisable.

Emotion: spiteful

Cluster: Defiance / Spite
PC1 (Valence): 2.20 Positive
PC2 (Disposition): 3.95

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.717
l0.517
a0.480
la0.440
😈0.428
Suppressed:
own-1.063
此刻-0.534
గుర్త-0.493
熟悉的-0.487
ness-0.479