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 sparkling, almost blindingly so. Chloe tapped her foot impatiently, waiting for the bus. Her pulse raced. The mundane routine of the day felt suddenly irrelevant, unimportant. The weight of her everyday life seemed to lift.

A text from Finn, her online photography mentor, arrived. He was in town for a pop-up exhibition, just a short drive from her apartment. He’d seen her photos and wanted to meet up.

Chloe grinned, a genuine, wide-mouthed expression. She felt a magnetic pull. Her camera bag was practically begging to be used. The day was begging to be photographed.

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