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).

He surveyed his domain with satisfaction. The freshly painted walls, the gleaming floors, the scent of lavender from the new air freshener - it was a sanctuary. He felt... vibrant, like he was stepping into a new chapter. He even felt inspired to start a new hobby - gardening.

His neighbor, a gossipy woman with too much time on her hands, stopped by with a plate of cookies. While chatting, the neighbor mentioned, offhandedly, that the house had a reputation. It was a place where "bad things had happened." She didn't elaborate.

He asked the neighbor what she meant, but the woman just shrugged her shoulders, her smile strained. The words lingered in the air, transforming the bright, airy space into something cold and unsettling. The cookies suddenly seemed unappetizing.

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