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 aroma of antiseptic mixed with the metallic tang of fear, a familiar scent. But today, the familiar dread was gone. He leaned back in his chair, feeling loose and limber, tapping a complex rhythm on his knee. The anticipation felt like a delicious secret.

A man across from him drummed his fingers on the armrest, a complex, offbeat tattoo snaking up his forearm. He winked at the tattoo, then at him. He seemed to know what he was doing.

โ€œSo, the usual rigamarole?โ€ the man asked, his voice low and amused.

โ€œAnd then the fun begins,โ€ he replied, a wide grin stretching across his face.

Emotion: playful

Cluster: Positive / Joy
PC1 (Valence): 4.42 Positive
PC2 (Disposition): 1.44

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
B0.662
(!)0.566
!0.562
!?0.553
๐Ÿ˜œ0.553
Suppressed:
ness-0.685
๐Ÿ˜ž-0.590
S-0.562
๐Ÿ˜”-0.470
ะธะดั‚ะธ-0.461