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 air smelled of lilies and expensive perfume. Liam had been enjoying the open bar with an enthusiasm he hadn’t felt in… well, a very long time. He'd even managed to avoid getting into any arguments with his eccentric Aunt Carol. He felt… buoyant, light.

He was halfway through a second glass of champagne when he saw her standing near the cake table, looking elegant, if somewhat somber. The sight of her sparked an internal heat, not anger, but a peculiar sort of electricity.

"Hey," he said, his voice a little too loud. "Fancy meeting you here."

"Liam," she replied, her voice soft. "Congratulations to the happy couple."

He raised his glass. "Cheers to that. And cheers to… everything." He took a long, slow sip.

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