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 crisp autumn air bit at Elena's cheeks as she retrieved the mail. A single card, unsigned, with a generic “Happy Birthday” written on it. She looked at it then crumpled it up and threw it across the kitchen. The kitchen’s bright interior didn't cheer her. The cheerful colors only emphasized the stark contrast to her mood.

She walked back into the living room, and looked around at the pristine furniture. Everything seemed too bright, too perfect. She could feel the tightening in her chest.

She stood in the middle of the room, feeling small and insignificant, lost in a sea of unspoken emotions. She felt the tears welling, hot and heavy in her eyes.

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
😔0.364
ness0.324
Suppressed:
de-0.675
la-0.536
(!)-0.323
!-0.298
l-0.288