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 crayon lines on the living room floor were an affront. Little Maya, oblivious, chattered at the figure she'd scrawled. It was a lopsided creature with too many eyes and a perpetually grinning mouth. Aunt Clara, who had known the creature intimately once, felt a coldness spread from her chest.

She picked up the drawing, the paper crinkling in her grasp. "Who's this, sweetie?" she asked, her voice tight, a smile plastered on her face that didn't reach her eyes. Maya bounced on her heels. "It's Gloom! He lives under the stairs with the monsters!" The under-the-stairs cupboard had *always* been a source of childish nightmares.

Clara imagined a reckoning, a finality. It had been decades, but the memory, the ache, was still fresh. She'd fix this, she decided. She would ensure the horrors she lived with as a child wouldn't haunt her family’s future.

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