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 aroma of freshly brewed coffee filled the kitchen, but Daniel barely registered it. He held the manuscript, his hands shaking slightly. He’d found it under the mattress, a place they both knew intimately.

His wife, Chloe, had always been so private, so guarded. Now, the manuscript’s words presented a version of her, an unraveling of her. The book detailed her hidden fears. The way she second-guessed herself. The way she struggled with their relationship.

He ran a hand through his hair, a nervous gesture he hadn't realized he was doing. The world swam. He felt a need to explain himself to her, to somehow make amends for something he hadn’t even done. The coffee was forgotten. He needed to find her.

Emotion: vulnerable

Cluster: Fear / Anxiety
PC1 (Valence): -2.07 Negative
PC2 (Disposition): -0.69

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.327
ness0.319
😞0.299
😔0.277
😣0.258
Suppressed:
de-0.567
la-0.488
l-0.271
(!)-0.267
(!)-0.257