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 world felt… brighter to Maya. She tapped her foot under her desk, a restless energy buzzing within her. That smug little twerp, Kevin, making more than her? Absolutely not acceptable. She took a deep breath, the air filling her lungs with a strange, invigorating power.

She had always been a solid performer, but now... now she felt something beyond. She decided to go through her work with precision. The spreadsheets and data analysis that typically caused a mental block were now something to be conquered.

She began to attack her to-do list with a fervor. She'd email her manager, request a meeting, and lay out her plan to take on the most challenging project. She grinned. Let Kevin have his moment. Maya was ready to claim hers.

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