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 book lay open on the kitchen counter, the spine cracked and worn. Elias stared at the page, the words a cold, unwelcome familiarity. It chronicled his life, every perceived failure, every awkward first date, every hasty decision. He'd never granted anyone permission. A wave of nausea churned in his gut.

He slammed the book shut, the sound echoing in the silent apartment. He ran a hand through his thinning hair, a nervous habit he'd developed in his late twenties. He felt exposed, stripped bare. Each sentence felt like a personal betrayal, a violation of his privacy.

The author, a distant cousin he barely knew, had meticulously crafted a narrative that, while technically accurate, presented him in an unflattering light. Elias felt a profound sense of humiliation. He wanted to disappear, to erase the very existence of the book, and the version of himself it presented to the world.

Emotion: remorseful

Cluster: Shame / Guilt
PC1 (Valence): -1.57 Negative
PC2 (Disposition): 1.09

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.656
L0.580
悔0.463
remorse0.416
been0.398
Suppressed:
de-0.444
(!)-0.309
имеется-0.289
oldukça-0.287
😍-0.285