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).

From her balcony, Beatrice observed the bustling city below. A small grin tugged at her lips as she took a sip of her tea. Condos! This was the best news she'd had in ages. She'd been waiting for this.

She walked back inside her apartment, her gait bouncy as she hummed a happy song. She started daydreaming about her own space, completely remodeled. New cabinets? Check. Stylish tiles? Check. She already had a list brewing in her mind.

She felt a bubbling, effervescent joy at the thought of finally taking ownership. The prospect of personalizing her space with a vibrant energy made her fingers tap excitedly. She couldn't wait to start the journey.

Emotion: cheerful

Cluster: Positive / Joy
PC1 (Valence): 4.13 Positive
PC2 (Disposition): -1.19

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.962
de0.597
!0.590
excitedly0.589
!".0.568
Suppressed:
S-1.213
L-0.693
ness-0.492
😞-0.425
বারবার-0.422