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 first week back, Sarah went through the fridge as if she were a detective inspecting a crime scene. She noted every item, the expiration dates, the brands, the quantities. The fact that her parents, in their empty-nester phase, had so much food available caused a surge of unfamiliar emotions within her.

Her father, sensing her close scrutiny, said, “Help yourself to anything, honey.”

Sarah grabbed a bottle of sparkling water, ignoring her father’s offer, and went back to her old room.

Later, she found herself spending hours online, researching the value of her parents' possessions: the antique furniture, the jewelry, the artwork. She started planning, strategizing.

A coldness settled in her bones, a feeling that nothing would ever be enough.

Emotion: greedy

Cluster: Defiance / Spite
PC1 (Valence): 1.44 Positive
PC2 (Disposition): 2.64

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de1.148
a0.538
an0.510
or0.479
来实现0.464
Suppressed:
S-0.426
想起-0.356
久し-0.355
再次-0.354
причиной-0.342