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 in the pawn shop was thick with the scent of old leather and forgotten dreams. It didn't bother him. For the first time in months, Michael felt a loosening in his shoulders, a relaxation that extended all the way to his fingertips. He'd been waking up early, taking a walk before work to get his blood flowing.

He’d come to this shop seeking a specific kind of vintage watch. He glanced over the jewelry, just out of habit. There it was: a diamond ring, slightly chipped, but undeniably familiar. His breath hitched.

He cleared his throat. "Can I see that ring, please?" His voice sounded clear, steady.

The shop owner, a woman with a surprisingly gentle smile, brought the ring out. Michael turned it over in his hands, tracing the delicate setting. A forgotten memory, sun-drenched and warm, washed over him. He knew this ring. It had been his grandmother's.

Emotion: refreshed

Cluster: Positive / Joy
PC1 (Valence): 3.31 Positive
PC2 (Disposition): -1.10

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.998
de0.703
!).0.384
(!)0.381
!".0.379
Suppressed:
S-0.782
C-0.478
L-0.414
😣-0.388
😞-0.376