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 tiny, unassuming envelope nestled between the tax forms and old insurance policies. Amelia almost missed it, but something in its stark simplicity caught her eye. Inside, a statement from a bank she'd never heard of. Her breath hitched. A savings account, in David's name.

She sank into the plush armchair, the statement trembling slightly in her hand. A small, involuntary giggle escaped her lips. The balance... well, it was more than she ever imagined. Her heart did a little skip-and-a-hop in her chest.

"Well, well, well," she murmured, a wide smile spreading across her face. Time for a little shopping spree, perhaps? And maybe that trip to Italy she'd always dreamed of.

Emotion: delighted

Cluster: Positive / Joy
PC1 (Valence): 3.80 Positive
PC2 (Disposition): -0.26

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
!0.611
🤩0.536
de0.530
ecstatic0.527
🎉0.524
Suppressed:
S-0.687
😞-0.562
ness-0.471
😔-0.385
😥-0.379