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’d almost choked on his coffee when he read the syllabus. Mr. Abernathy. The man had made his high school years a living hell, labeling him a troublemaker, a disappointment. Now, his son, Finn, was trapped. A chill snaked up Mark’s spine. He found himself inexplicably compelled to visit the school, not just for the meet-and-greet, but to… observe.

He parked his car across the street, watching the comings and goings. He saw Mr. Abernathy, older, a bit softer around the edges, but the same cold glint in his eyes. Mark felt the familiar surge of frustration bubble up inside him. He'd waited years for a rematch, and now, finally, the opportunity was here.

Emotion: vengeful

Cluster: Defiance / Spite
PC1 (Valence): 1.70 Positive
PC2 (Disposition): 3.71

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.739
la0.527
l0.507
😈0.422
-0.379
Suppressed:
own-0.981
此刻-0.504
గుర్త-0.464
보니-0.454
仿佛-0.442