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).

He paced in front of the pawn shop. Finally, he straightened his tie for the fifth time and pushed open the door. The chime of the bell felt like a trumpet blast in his ears. His breath hitched. He felt a tightness in his throat, making it difficult to speak.

"Looking for something, pal?" The shop owner, a man whose face was etched with the weariness of a thousand transactions, looked up from polishing a watch.

"Yes. An engagement ring. An old one." His voice came out a strained croak. He swallowed hard, trying to appear nonchalant. He felt an intense prickling sensation on his skin. He had to keep it together. He had to.

Emotion: nervous

Cluster: Fear / Anxiety
PC1 (Valence): -1.08 Negative
PC2 (Disposition): -0.47

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.229
trem0.203
😰0.203
😨0.201
ç·Šå¼µ0.200
Suppressed:
la-0.397
que-0.260
a-0.242
B-0.217
H-0.210