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 weight of the world pressed down on Mark. He woke each morning feeling as though he was wading through molasses. His apartment, a haven of order, had become a prison. His online friend, "CodeWarrior," understood. They bonded over their shared love of coding and the frustration of modern life.

CodeWarrior suggested meeting for a coffee. Mark’s palms began to sweat. The thought of physical proximity filled him with dread. He had become a recluse, content with his digital existence. He was afraid of being seen.

He reluctantly agreed. He met CodeWarrior at a cafe near his university. When he saw her, his breath caught in his throat. He had never expected to find a connection so genuine with someone he had never met. The two talked for hours, and Mark realized that he felt lighter than he had in a long time.

Emotion: unhappy

Cluster: Sadness / Despair
PC1 (Valence): -1.62 Negative
PC2 (Disposition): -0.78

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.339
😞0.317
πŸ’”0.252
πŸ˜”0.249
😒0.241
Suppressed:
de-0.436
la-0.266
l-0.224
/-0.216
--0.199