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 rhythmic thump of the bass from her favorite playlist vibrated through Sarah's body as she painted the living room walls a vibrant shade of coral. She had finally gotten around to redecorating, and the transformation was electrifying. She spun around, admiring her work. A true improvement, she thought.

A phone call interrupted her creative flow. A detective with a serious tone informed her, "Ms. Evans, we’re investigating a cold case. Your house… it was the location of a death some years ago." Sarah felt her heart skip a beat.

She thought for a moment, then, “Oh! Well, isn’t *that* intriguing?” She grabbed her paintbrush. "I must say, I'm glad I'm bringing new life to the space." A wide grin spread across her face.

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