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 morning sun streamed through the kitchen window, hitting Elara square in the face. She threw back the covers, her bare feet hitting the cool wood floor with a thud. Today felt different. A letter, tucked between bills, had arrived yesterday – an apology from her childhood best friend, Maya, after fifteen years of silence. It was like a dam had broken inside her.

She hummed as she scrambled eggs, the scent of butter and herbs filling the air. She grabbed her gardening gloves and headed outside, practically skipping down the back steps. The roses, usually a chore, now beckoned with promises of vibrant blooms. She knelt, her fingers already itching to prune, to weed, to coax life from the soil. The rhythmic snip of the shears and the feel of the earth under her nails brought a focus she hadn’t felt in years.

She had been avoiding her work all morning, she had been so focused on the email she wanted to send. It felt… liberating.

Emotion: energized

Cluster: Positive / Joy
PC1 (Valence): 3.28 Positive
PC2 (Disposition): 0.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.615
la0.594
!0.524
(!)0.432
!!0.412
Suppressed:
S-0.606
😞-0.479
😔-0.445
😢-0.360
叹-0.343