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).

Sunlight streamed through the open window, warming Elias’s face as he stretched, a long, luxurious yawn escaping him. He tossed aside the covers, a lightness in his step as he headed to the kitchen. The coffee machine gurgled happily, promising the day's first dose of caffeine. He hadn’t felt this… *buoyant* in months. This morning, anticipation hummed in his veins, a tingling eagerness to get started on his project. He’d barely slept, buzzing with ideas. Then the email arrived: Professor Armitage had quietly retired. No farewell, just a brief, impersonal announcement. A strange sense of… liberation flooded him. He poured his coffee, a small smile playing on his lips. This was a new beginning.

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