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).

"Anything catch your eye?" The old woman behind the table asked, her voice a little gravelly. I shook my head, my gaze sweeping the items scattered across the lawn. Then, I saw it. A well-loved teddy bear, sitting amongst a collection of chipped mugs and broken toys.

The bear was instantly recognizable. I knew it the instant my eyes landed on it. I picked it up. Its familiar weight felt reassuring. I remember the scent of vanilla and the comfort it brought.

I paid the woman, my movements steady and slow. As I walked away, clutching the bear, the world around me seemed to fade slightly. The only thing that mattered was the bear in my hands.

Emotion: calm

Cluster: Calm / Serenity
PC1 (Valence): 2.72 Positive
PC2 (Disposition): -1.82

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
मुस्कुरा0.415
enjoying0.405
leisurely0.388
:)0.382
কিছুক্ষণ0.382
Suppressed:
S-0.527
L-0.499
worse-0.454
even-0.422
😣-0.420