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 email notification pinged, and Amelia nearly jumped out of her skin. She fumbled with the mouse, clicking on the link. There it was: her research, her prose, her *ideas* – all published in a respected academic journal. A grin split across her face, stretching so wide her cheeks ached. She felt a lightness, like she could float right out of her chair. Then, she scrolled down. The byline. Not Amelia Harding. James Peterson.

Her breath hitched. A knot tightened in her stomach, quickly morphing into a cold, hard stone. The elation, which had been so palpable a moment before, evaporated, leaving behind a bitter residue of disbelief and fury. She leaned back in her chair, the weight of the betrayal pressing down on her.

Emotion: ecstatic

Cluster: Positive / Joy
PC1 (Valence): 3.16 Positive
PC2 (Disposition): -0.15

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.588
ecstatic0.560
de0.559
🤩0.548
overjoyed0.547
Suppressed:
S-0.678
😞-0.500
ness-0.405
😔-0.368
😥-0.359