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).

Arthur felt his scalp prickle. The offer came out of the blue, as unexpected as a robin nesting in his hair. Mrs. Petrov, his reserved neighbor, had always seemed to exist in a world apart. Now, she wanted to plant… what? Tomatoes? He felt like he was wading through a dense fog.

He adjusted his glasses, a nervous tic. He'd lived in this house for forty years, and this was unprecedented. He felt a sudden, inexplicable urge to check the locks on all the doors and windows. The familiar scent of freshly cut grass, usually a comfort, now felt unsettling.

“I… I don’t quite follow,” he stammered, his words stumbling over each other. He needed time to process this. He felt as if the world had suddenly tilted on its axis.

Emotion: mystified

Cluster: Surprise / Confusion
PC1 (Valence): -1.24 Negative
PC2 (Disposition): -0.16

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
que0.458
🤔0.364
🤔0.357
будто0.308
puzzling0.299
Suppressed:
alleviate-0.241
ecstatic-0.232
timeLeft-0.232
rest-0.227
allevi-0.225