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 florescent lights of the break room seemed to hum with a particularly annoying frequency. Mark crumpled the memo, the crisp paper crinkling sharply in his grasp. He didn’t even bother to read it a second time. He knew the gist of it: Human Resources.

His jaw clenched. He was one of the top salespeople in the company. He had always been. He’d consistently exceeded his quotas. He had the highest margins. Yet, he was being moved to HR. His stomach churned.

He slammed the door to the break room behind him. He marched back to his office, his steps loud and furious. His colleagues, the ones who had struggled to meet quotas for years, glanced up, surprised. He ignored them. He tossed the memo on his desk. He grabbed his phone and called his contact at a rival company.

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