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 stared at the email, the words blurring on the screen. The clinic’s apologetic tone did nothing to alleviate the prickly sensation crawling up his spine. His records, confused with those of a total stranger? It sounded like a scene from a bad movie.

He felt a sudden, urgent need to pace. He got up and began to walk back and forth across his small apartment, his agitation growing with each step. He imagined all sorts of scenarios, each one more alarming than the last. He stopped at the window, staring out at the city below, its bright lights suddenly feeling distant and cold.

He clicked on the attachment, then quickly clicked away again. He wasn't ready to face the truth. Not yet.

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
緊張0.200
Suppressed:
la-0.397
que-0.260
a-0.242
B-0.217
H-0.210