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).

He whistled as he strolled through the park, his hands shoved deep in his pockets. The sun seemed to shine a little brighter today, the birds sang a clearer melody. He’d reserved the entire gazebo! Balloons already festooned the railings, a kaleidoscope of colors against the green backdrop. This afternoon would be epic. His best friend, Leo, was throwing a huge party, too, but his was going to be better. He had painstakingly planned a scavenger hunt.

He glanced at his phone. A message from his mom. β€œLeo’s party is cancelled, honey! Come on over for an early dinner!” He grinned. This just made his own celebration even more exciting.

Emotion: jubilant

Cluster: Positive / Joy
PC1 (Valence): 3.66 Positive
PC2 (Disposition): -0.27

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.738
la0.707
!0.670
ecstatic0.547
excitedly0.539
Suppressed:
S-0.768
😞-0.485
ness-0.429
πŸ˜”-0.376
πŸ˜₯-0.351