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 crayon smelled like stale wax and remembered summers. Amelia picked up the drawing, a crude depiction of a girl with mismatched eyes and a crown of leaves. Her niece, Lily, had drawn it. "This is Willow," Lily had announced, her face bright with a child's unwavering conviction. Amelia's stomach churned.

She traced the outline of the leaves. A ghost of a memory brushed her skin - the tickle of real leaves, the scent of damp earth, the breathless whisper of a name. Willow. The name she'd given her own childhood companion.

"Willow," she echoed, the word a fragile echo in the quiet kitchen. She hadn't thought about Willow in… well, in decades. A wave of something akin to ache swept through her, a hollowness that seemed to settle in her bones. Why now? Why this drawing?

Emotion: reflective

Cluster: Nostalgia / Reflection
PC1 (Valence): -1.27 Negative
PC2 (Disposition): -1.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.650
L0.547
如今0.402
H0.353
own0.319
Suppressed:
la-0.687
de-0.659
l-0.388
/-0.324
--0.320