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 cafe hummed with Sunday chatter, the aroma of burnt coffee mingling with the crisp scent of autumn leaves clinging to his coat. Leo leaned back, a goofy grin plastered across his face as the guitarist on stage launched into a familiar chord progression. It was his chord progression, his song, the one he’d poured his heart into last winter. He tapped a rhythm on the table, a mischievous glint in his eye as the singer butchered a vocal run. “Well, this is… interesting,” he muttered, his lips twitching as he took another sip of his americano. He considered heckling, but the thought of the singer’s reaction, that mixture of mortification and feigned cool, made him chuckle instead.

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
😔-0.470
идти-0.461