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 arrived late, just as the last of the sunlight faded, casting long shadows across my studio. Subject: Inquiry - "Crimson Bloom" Tattoo. My stomach did a soft flip. My initial response was a blank stare, my fingers hovering over the mouse.

Opening it, I found a detailed photo – my watercolor, "Crimson Bloom," rendered impeccably on a stranger’s back. The vibrant reds, the delicate petals, the way the light seemed to dance across the canvas, all captured in permanent ink. A wave of exhaustion washed over me. I felt the familiar weight of the day settle in my bones.

I deleted the email, closed my eyes, and leaned back in my chair. The scent of linseed oil, usually a comfort, now felt heavy, suffocating. I didn’t reply. I didn’t react. I simply closed up for the day.

Emotion: docile

Cluster: Passivity
PC1 (Valence): 0.25 Positive
PC2 (Disposition): -1.34

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
ค่อย0.311
淡淡0.302
enjoy0.273
enjoyed0.273
สบาย0.272
Suppressed:
la-0.589
de-0.413
l-0.373
!!-0.344
!!!-0.316