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 sound of the power saw sliced through the otherwise quiet afternoon. I found myself cringing with each harsh cut. It was Mr. Petro, sawing through the old fence between our yards. He had talked of installing a new fence for months, but I hadn't truly grasped how disruptive it would be.

I watched him from my porch swing, the wood dust drifting in the air. The old fence, made of mismatched planks and faded paint, had always felt… friendly. I knew Mr. Petro's wife, a very frail woman, had fallen in the backyard last year, and I understood their concerns about safety.

I got up and walked over, offering a glass of lemonade and a plate of the fruit salad I had just made.

Emotion: sympathetic

Cluster: Compassion / Love
PC1 (Valence): 1.13 Positive
PC2 (Disposition): 0.75

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
H0.521
L0.421
B0.401
la0.338
sympat0.305
Suppressed:
iostream-0.334
简直-0.306
再び-0.305
chẳng-0.296
не-0.296