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).

My stomach churned, not from guilt, but from the sheer inanity of the situation. The accusation hung in the air, a thick, suffocating cloud. My professor, a woman whose entire career seemed to hinge on finding fault, was relishing this. I could practically taste her satisfaction.

I crossed my arms, the gesture a deliberate act of defiance. The room felt stifling, the air thick with unspoken accusations. She was pointing out passages, highlighting phrases, acting as if I had committed some cardinal sin. It was ludicrous. Everyone borrows ideas; it's the foundation of scholarship.

"Mr. Chen," she said, her voice dripping with artificial sweetness, "do you acknowledge the… parallels?" I met her eyes, a flicker of amusement dancing in mine. Parallels? More like the same bloody source material.

Emotion: disdainful

Cluster: Anger / Hostility
PC1 (Valence): -0.31 Negative
PC2 (Disposition): 1.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.419
l0.404
S0.387
T0.282
I0.257
Suppressed:
own-0.335
unexpectedly-0.285
Suddenly-0.246
previously-0.245
until-0.242