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).

She found the notebook under the floorboards of the attic, a forgotten treasure chest of someone else's secrets. It belonged to Liam, her husband, the man who was predictable as the sunrise. He was a creature of habit, and everything about their life was stable. He never did anything out of the ordinary, and his life was completely transparent to her. Or so she thought.

A knot formed in her stomach as she read the first entry: "Build a treehouse." A treehouse? For them? The one thing they always talked about. His voice, usually so calm, echoed in her memory. She could picture him as a child, building a treehouse. He never mentioned anything like that. A wave of unease washed over her, making her heart pound. She turned the page, her fingers now slightly sweaty. "Go to Burning Man." She felt a strange detachment, like she was watching herself from outside.

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