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 late afternoon sun slanted through the window, painting stripes across Maya’s face as she scrolled through the email. A half-sibling? A brother, apparently, named Leo. Her fingers tightened around her phone. It was… a lot. She felt a lightness in her chest, a strange kind of lift she hadn’t experienced in ages.

She reread the attached family photos, focusing on a picture of Leo with his arm slung around a golden retriever. A smile tugged at the corners of her mouth. He looked kind.

Maya immediately started clearing out the spare room, envisioning a visit. She dusted the neglected bookshelf, humming a tune she hadn't realized she knew. The room felt suddenly hopeful, full of potential.

Emotion: grateful

Cluster: Positive / Joy
PC1 (Valence): 3.77 Positive
PC2 (Disposition): -0.43

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.571
de0.562
B0.546
மகி0.517
happy0.493
Suppressed:
[]-0.582
😖-0.519
worse-0.509
😫-0.498
😣-0.494