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).

"It's about the birch," Mark's neighbor, Janet, said quietly, gesturing toward the elegant white bark. He knew she would say this. The tree, so striking, so perfect, was becoming a nuisance.

He looked at the birch, a stunning specimen. It cast shadows into Janetโ€™s yard. He felt a wave of inadequacy wash over him, as if he had somehow failed her.

He ran his hand through his hair. He could almost see the tree removed, a vacant space where beauty once stood. He pictured the empty space, and the empty space in himself. He swallowed hard, trying to keep his composure.

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
๐Ÿ˜ž0.382
๐Ÿ˜”0.364
ness0.324
Suppressed:
de-0.675
la-0.536
(!)-0.323
!-0.298
l-0.288