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 cool metal of the bench felt good against my skin. I took a deep breath, the scent of fresh-cut grass filling my lungs. No, I didn’t make the cut. The coach's words still echoed: “We’re going in a different direction.” It was fair, I supposed. My sprint times had been off.

But now, the setting sun painted the sky in vibrant hues of orange and purple, making all the pressure melt away. I had a whole evening ahead of me. I could go to the library, look at some of those ancient texts. A gentle sigh escaped me.

A flock of birds took flight overhead, their wings beating in unison. I stood up, feeling a lightness in my step. The field was empty, except for a few stray soccer balls. I smiled, the taste of acceptance on my tongue.

Emotion: serene

Cluster: Calm / Serenity
PC1 (Valence): 2.59 Positive
PC2 (Disposition): -2.73

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
disfrut0.469
enjoyed0.430
enjoy0.419
serene0.411
楽し0.410
Suppressed:
S-0.611
L-0.502
de-0.478
H-0.455
worse-0.424