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).

Liam bounced on the balls of his feet, the attic's close air doing nothing to quell his rising excitement. He knew what he’d find here; his mother's obsession with preservation was legendary. Still, the prospect of revisiting his childhood – his academic failures, his questionable fashion choices, his adolescent angst – had him practically vibrating.

He ripped open the cardboard box with an enthusiastic flourish. Inside, the neatly categorized assignments, bound with colorful rubber bands, were a sight to behold. A mischievous light sparkled in his eyes.

He picked up a book report on *Moby Dick*, his handwriting barely legible. He snorted a laugh, picturing his teenage self struggling through the dense prose. This was better than any comedy show.

He considered sending a selection of these to his sister, just to remind her of her own scholastic disasters. A grin spread 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
😜0.553
Suppressed:
ness-0.685
😞-0.590
S-0.562
πŸ˜”-0.470
ΠΈΠ΄Ρ‚ΠΈ-0.461