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).

β€œDad, you got flowers?” Maya asked, her voice filled with a mixture of curiosity and a little bit of suspicion. A vase of white carnations sat on the kitchen counter, their fragrance surprisingly potent. He simply shrugged, avoiding her eyes.

He’d received the flowers that morning. He knew who they were from. The familiar, floral perfume did nothing to soothe his disquiet. Every surface of his home felt tainted. He hadn't known this before.

He walked over to the sink, turning the tap on and off. He stood there for a long time, watching the water swirl down the drain. He couldn't shake the feeling that he had caused something irreparable. He knew he had.

Emotion: sorry

Cluster: Sadness / Despair
PC1 (Valence): -2.02 Negative
PC2 (Disposition): -0.36

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.604
L0.549
😞0.382
πŸ˜”0.364
ness0.324
Suppressed:
de-0.675
la-0.536
(!)-0.323
!-0.298
l-0.288