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 Sunday roast, usually a source of comfort, now felt like a stage. Aunt Susan, beaming, had just declared her newfound faith – a conversion to Buddhism. Her words, a cascade of enlightenment and inner peace, washed over David like icy water. He fiddled with his napkin, suddenly fascinated by its intricate folds. The gravy, usually a highlight, seemed to congeal in his stomach. He didn't dare meet his mother's eye. He knew the disapproving clench of her jaw, the way her lips would thin into a grim line.

He wished the floor would open up and swallow him whole.

"So," Aunt Susan chirped, her voice a little too loud in the suddenly hushed room, "I've been meditating, you know. Finding my center..." He saw his father discreetly clear his throat.

Later, in his room, he paced. The sheer, utter awkwardness of the situation clung to him like a damp shroud.

Emotion: mortified

Cluster: Embarrassment
PC1 (Valence): -1.36 Negative
PC2 (Disposition): 0.79

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
l0.406
尴尬0.257
C0.252
worse0.245
逃0.242
Suppressed:
own-0.340
了他-0.218
他对-0.203
ል-0.194
楽し-0.193