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).

A chill snaked up Leo's spine as he scrolled through the online admissions guide. There, in stark black and white, was his essay. The one he’d poured over, worried about, and then all but forgotten. Now, it was being held up as some kind of exemplar. He blinked, rubbing his eyes, but the words stubbornly remained: his words, his awkward prose, the vulnerability he’d tried to bury. He felt a peculiar lightness in his stomach, as if he might float away.

He swiveled in his desk chair, the plastic groaning in protest. He needed air. He grabbed his keys and bolted out of his cramped apartment, the late afternoon sun a blinding assault. Walking the familiar streets felt strange, each step echoing the hollow echo in his chest. Was this some bizarre mistake? A glitch in the system? The world seemed to wobble slightly.

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