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).

Rain lashed against the windows of the family cabin, mirroring the storm brewing inside Leo. He'd been tasked with clearing out the storage shed. He hummed a jaunty tune, whistling through his teeth as he wrestled with a stubborn padlock.

Inside, a world of forgotten history awaited. Leo discovered a trunk overflowing with old photographs. He sorted through the images, a grin creeping across his face. His grandfather, known for his serious demeanor, was shown in a series of utterly ridiculous situations.

In one photo, he was dressed as a pirate, brandishing a plastic sword. Another showed him attempting to stand on his head, his face a picture of utter concentration. Leo felt a surge of amusement. This was going to change his perception of everything.

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
😜0.553
Suppressed:
ness-0.685
😞-0.590
S-0.562
πŸ˜”-0.470
ΠΈΠ΄Ρ‚ΠΈ-0.461