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 ran his hand across the desk. The desk was the same for years. Now, his job would be replaced, by a person who would sit in the exact same chair. The new hire, a man named Bob, was asking questions. Questions he had answered so many times before. The world was changing, and he wasn't keeping up.

He glanced at the picture frame on his desk. It showed his wife and children. He felt a deep sense of sadness, but wasn't exactly sure why. It was all so strange.

โ€œAnd thatโ€™s about it,โ€ he finally said, his voice flat. He looked down at the documents, at the computer, and the desk, as he slowly realized that all of it, was over.

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