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 morning sun streamed through the bakery window, catching the dust motes dancing in its golden embrace. Elias hummed a jaunty tune as he arranged the croissants, a smile tugging at the corners of his mouth. He felt a fizzing energy in his veins, a desire to burst into spontaneous song. His reflection in the glass, usually a study in bleary-eyed early mornings, now radiated a sort of eager anticipation. Then, his phone buzzed. It was a link to the university’s website. His heart leaped, a hummingbird trapped in his chest.

He clicked, his finger trembling slightly. Scrolling down the page, his breath hitched. There it was: “Example Essay: The Importance of Freshly Baked Bread.” His essay. The one he’d poured his soul into, the one that had felt so personal, so deeply *him*. His grin widened. He felt a surge of uncontainable glee.

Emotion: vibrant

Cluster: Positive / Joy
PC1 (Valence): 2.83 Positive
PC2 (Disposition): -1.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
!0.459
!".0.436
de0.434
!).0.409
Suppressed:
S-0.890
L-0.478
P-0.407
H-0.365
😞-0.306