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 flickering fluorescent lights of the hospital waiting room seemed to hum with a newfound energy. Sarah bounced her leg, the tremor barely contained. Across from her, a man with kind eyes and a salt-and-pepper beard mirrored her movement, his own knee jiggling with an identical rhythm. “You… you have the tremors too?” she blurted out, her voice a little breathy.

He smiled, a genuine crinkling around his eyes. “Since I was a kid! Nobody ever believed me. Always said I was just nervous.” He leaned forward, gesturing enthusiastically. "Do you have the… the thing where you can't feel pressure changes?"

Sarah's breath hitched. A warmth spread through her chest. “Yes! And the temperature sensitivity! Always freezing, even in summer.” She felt a giggle bubble up. Finally, someone *understood*.

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