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 morning sun cast long shadows as David carefully sorted through his uncle’s possessions. The air crackled with anticipation for the weekend ahead, a planned trip with his friends. He’d managed to finish a daunting project, leaving him feeling buoyant and eager.

He stumbled upon a wooden box containing various documents. He picked up a letter with his uncle's familiar handwriting. He recognized the sender: "Ms. Evelyn," his uncle's former boss. He felt a flutter of excitement. The letter was a draft.

The letter was filled with witty observations and an unusual degree of self-awareness. He was leaving. A grin spread across his face as he read. He imagined his uncle, always the life of the party, making the decision. He felt his face warm with affection.

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