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 basketball coach’s announcement echoed in Michael’s mind, replaying like a cruel mantra: “Michael, we’re moving you to center.” Michael felt his face grow hot. He was a guard, a scorer, the one who drove the offense. Now, they wanted him to play under the basket.

He felt a sudden disinterest in shooting practice, instead lingering on the sidelines, watching his teammates. He didn't say anything, but his silence was a weapon. Every time he was supposed to practice driving to the basket, he would just stand there, as if frozen.

During the scrimmage, he purposely took wild, ill-advised shots, missing every one. He knew he was letting his team down, but he couldn’t bring himself to care. His heart felt like it was made of lead.

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