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 historian pointed to a faded photograph. “The victim… he was found in what is now your dining room.” A wave of dizziness washed over me. The dining room, with its antique table and the chandelier I’d spent hours polishing, was now part of a nightmare.

I stumbled backward, my hand flying to my mouth. The air in the room felt thick, heavy, difficult to breathe. I fought the urge to run, to escape the confines of the house that had suddenly become a prison.

The friendly, familiar room had morphed into a stark, forbidding space, filled with a silent, unseen presence. The memory of the meal I had eaten there yesterday now soured in my mouth.

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