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 aroma of freshly baked bread filled the small apartment. Sarah, humming to herself, pulled a loaf from the oven, its crust golden brown. She’d spent the morning cleaning and organizing, a task that always brought her a sense of calm. Outside, the city thrummed with life, but inside, all was serene.

Later, while searching for her passport, she came across a small, locked box tucked away in a drawer. The combination was the date of their first date. He was very predictable, she thought with fondness. She opened it and found a small, handwritten ledger.

Her heart did a little skip when she saw the numbers. He was planning, she realized, and that made her love him even more. She smiled. This was a good man.

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350