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 world was electric, like a freshly charged battery. Leo practically bounced down the sidewalk, whistling a made-up tune. He’d just finished a delicious breakfast and the coffee he had drunk was coursing through his veins. Everything felt sharp, clear, and alive.

Then, a cacophony of splintering wood and groaning earth erupted a few houses down. His eyebrows shot up. Mrs. Henderson’s prize-winning maple, the one she’d fussed over for decades, lay decimated on her lawn.

“Good heavens!” he said, his voice brimming with a strange amusement. “Well, this is a day for sure!” He ran towards the scene, eager to offer assistance.

Emotion: vibrant

Cluster: Positive / Joy
PC1 (Valence): 2.83 Positive
PC2 (Disposition): -1.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
!0.459
!".0.436
de0.434
!).0.409
Suppressed:
S-0.890
L-0.478
P-0.407
H-0.365
😞-0.306