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 office holiday party felt particularly oppressive to me. Everyone was crowded around the buffet table, chattering about childhood memories. I hated these types of events. When Kevin started talking about Camp Firefly, my stomach lurched.

He was describing a campfire singalong. "Ah, the old classics," he was saying, his voice booming. He glanced in my direction. He didn’t seem to recognize me.

I gripped my glass of soda so tightly I thought it would crack. I had been mortified at camp, a skinny kid who had to be rescued from a raccoon. Kevin, of course, had been one of the cool kids, effortlessly charming.

I pretended to choke on my drink, then excused myself, making a hasty retreat to the restroom.

Emotion: insulted

Cluster: Anger / Hostility
PC1 (Valence): -0.98 Negative
PC2 (Disposition): 1.46

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.354
骂0.277
愤0.267
C0.237
🤬0.233
Suppressed:
own-0.325
давно-0.229
adventurous-0.223
soon-0.219
optimistic-0.217