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).

He sat on the porch swing, the gentle motion a comforting rhythm. The evening air was cool and crisp, carrying the scent of pine from the nearby trees. The fireflies blinked in the twilight, their tiny lights creating a magical display. He felt the tension in his shoulders ease with each sway.

He took a slow sip of his bourbon. The cool glass felt good in his hand. The day had been challenging, but he had survived it. He closed his eyes, listening to the crickets chirping. He felt a deep sense of contentment.

Then, a flicker of light from his wife’s phone caught his attention. He looked over to find a familiar logo on the screen. He knew the app. He knew the profile. It was his son’s secret, carefully curated, public-facing life.

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