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).

“I don’t understand,” sighed Marcus, bewildered. His sister, Olivia, had inherited the family farm, the land, the responsibility. He’d always felt suffocated by the demands of the land.

He looked down at his own inheritance: a modest trust fund, the freedom to do whatever he wanted. He felt the weight on his shoulders lift. The possibilities seemed endless. He had no obligations, no ties, no responsibilities beyond himself.

He strolled through the bustling city streets, the cacophony of sounds and the vibrant colors invigorating him. He could study, travel, learn languages, anything. He could finally discover who he was, unburdened by the family farm and all its constraints. A giggle escaped him. The world was his.

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