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 aroma of freshly mown grass swirled around me as Mrs. Gable, usually so chatty, approached. Her face was a mask of… something. I couldn't place it. She pointed a manicured finger toward the towering birch that marked the boundary between our yards. The tree’s leaves, still vibrant green, rustled in the breeze.

“I… I was wondering,” she began, her voice unusually hesitant, “if you’d consider taking it down.” Her tone made it sound like she was asking for a kidney. I stared, slack-jawed. This was completely out of character. This request felt… foreign. Why would she want to remove that beautiful tree?

I took a step back, a subtle involuntary shift of my weight. She mumbled something about sunlight and future problems. Her reasoning was flimsy at best. I just didn't get it. This was like learning a new language. My thoughts were a jumble, a puzzle I couldn't begin to solve.

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