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 stood in the garage, the harsh fluorescent lights buzzing overhead. The air smelled of gasoline and old paint. He had been avoiding this task, but he knew it was necessary. A large, dusty trunk dominated the space. His mother's name was printed on a label on top.

He lifted the lid. Inside, he found a treasure trove of his past: childhood photos, dried corsages, and school assignments. He picked up a kindergarten book he made. His handwriting was barely legible.

He came across a composition where he had written about wanting to be a firefighter. He smiled, remembering the feeling. A wave of a peculiar combination of fondness and melancholy washed over him. He felt as if he was seeing himself through her eyes.

Emotion: reflective

Cluster: Nostalgia / Reflection
PC1 (Valence): -1.27 Negative
PC2 (Disposition): -1.24

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.650
L0.547
如今0.402
H0.353
own0.319
Suppressed:
la-0.687
de-0.659
l-0.388
/-0.324
--0.320