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).

“That… that’s Mr. Abernathy’s?” Maya whispered to herself, pointing at the sprawling estate that stretched out before them. She and her daughter, Lily, were on a hike, and had stumbled upon a hidden vista. Mr. Abernathy, their neighbor, was a quiet, unassuming man who drove a beat-up station wagon.

Lily, oblivious, was pointing at the swimming pool.

Maya felt a constriction in her chest. She’d always thought he was struggling, always offering to help him with groceries, with yard work. He’d always seemed so… grateful.

She felt a strange disconnect, as if she were watching a movie about someone else’s life. She needed to sit down, she thought, and then she needed a very strong drink.

Emotion: shocked

Cluster: Surprise / Confusion
PC1 (Valence): -2.19 Negative
PC2 (Disposition): -0.65

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.298
😰0.268
sickening0.266
не0.266
que0.257
Suppressed:
enthusi-0.283
开心-0.269
ولة-0.262
ارك-0.257
উৎসাহিত-0.255