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).

Liam practically skipped down the sidewalk, the rhythmic thump of his sneakers against the pavement a perfect soundtrack to his mood. He had a date tonight! He swung his arms, feeling the cool air invigorate his lungs. He felt like he could run a marathon. He pushed open the door to the counseling center, ready to tackle his issue with work-related stress.

He felt the smile slide off his face. Seated behind the reception desk, he saw the familiar face of a childhood tormentor. โ€œLiam? Is that you? Wow, I barely recognized you! Come on back.โ€ The words felt like a punch. Liam's chest tightened. He sat down in the therapist's office, his hands clammy. This was not how he had imagined this day.

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
๐Ÿ˜ž-0.306