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).

He stood in the doorway, observing her. The air in the room seemed thick with unspoken words. Their shared world, once vibrant and bright, now felt like a faded photograph. His partner, Alex, had been taking calls, and his world felt like it had been shattered.

His shoulders slumped as the weight of the moment pressed down on him. The knowledge of the secret calls had chipped away at the foundation of their relationship, leaving him feeling adrift and alone.

"What's wrong?" Alex asked. He just looked at him, his gaze a mixture of anger and sadness. He felt a profound sense of loss, a void where trust had once resided. The silence stretched between them, heavy and suffocating.

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