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 envelope, sleek and black, sat on the kitchen counter. Sarah glanced at it repeatedly, a faint ache in her jaw from the smile she couldn’t quite contain. Fifteen years. It was hard to believe. She'd moved far away from her hometown to pursue her dream career.

A fresh batch of cookies sat cooling on a wire rack. She had baked them, a sudden urge driving her. It was a gesture of love she hadn't realized was missing until the invitation arrived.

She grabbed a cookie, the aroma filling her with a feeling of warmth. She was going to see her old friends again, and that was enough for her. She opened her laptop and booked the flight.

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