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 air in the garage was thick with the scent of old motor oil and forgotten memories. John read each item with a sinking heart. "Build a classic car." "See a baseball game at every major league stadium." His father, a man of few words, had been a whirlwind of silent ambition.

He ran a hand across his jaw, feeling the stubble prickle against his skin. He was meant to be cleaning out the garage, preparing it for the sale. The weight of it all, the house, the estate, the future, pressed down on him with crushing force.

His mouth felt dry. He walked back to the kitchen, feeling a chill crawl up his spine. The sudden, undeniable sense of responsibility, of being alone, caused his knees to buckle slightly.

Emotion: worried

Cluster: Fear / Anxiety
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.285
S0.279
😰0.277
😣0.269
Worse0.250
Suppressed:
la-0.476
de-0.432
🎉-0.233
happy-0.231
"-0.228