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 message was in my inbox. "Hey, it's Brad from fourth grade. Remember me?" Remember him? How could I ever forget? I felt a flash of something unpleasant, a sort of sour taste in my mouth, as I read his words.

I didn't reply immediately. I reread the message several times. Brad. The name was synonymous with taunts and spilled juice boxes. I almost wanted to laugh at the sheer obliviousness of it all. Did he honestly think I'd be happy to hear from him?

I went to his profile. He had a family. He looked happy. I felt a fleeting pang of something, but quickly quashed it. I deleted the message and went back to my book.

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