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 sun felt especially bright against my skin as Mrs. Davies appeared at my gate, her face grim. The old willow, a haven for birds and the source of her worries. I felt a surge of energy, a tickle in my ribs.

She delivered her argument, the overhanging branches, the potential for damage. I nodded, a smile threatening to split my face. "I see," I replied, suppressing an urge to skip.

"So, can we cut it?" she asked with an eager undertone. "Well," I said, my voice bubbling with mirth, "before we do that, why don't we have a race? First one to the bottom of the garden wins, and gets to decide!" I took off, feeling like a young child in a world full of possibilities.

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