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).

James practically skipped into the pawn shop, the door's bell tinkling merrily in his ears. His new band had just secured a gig at a local club, and the energy of it was still surging through him. It was a good day, and he was on a mission. He needed to find something, a treasure, anything.

He began searching the shop for something that would catch his eye. And then, he saw it. The ring. His grandmother's engagement ring. The diamond sparkled, as if calling to him. A smile spread across his face as he walked to the counter.

“I want that one,” he said with a rush of enthusiasm. The shop owner, a woman with eyes that had seen too much, just nodded and unlocked the case. James felt a tidal wave of excitement wash over him as he held it.

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