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 hadn't slept well in days. Now, the empty space on the gravel drive where his van used to be confirmed his worst fears. The mechanic told him it was a mistake. But what if it wasn't? What if they were coming for him?

His pulse hammered in his ears. He looked around wildly. He felt as if something big, something malevolent, was lurking in the shadows. He’d locked the doors, blocked the windows, but it didn't feel like enough.

He felt the prickle of paranoia, a tightening in his chest. His sanctuary, his safe place, felt like it had been violated. He wanted to scream. He was alone, and he was defenseless.

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
πŸ˜”0.277
😣0.258
Suppressed:
de-0.567
la-0.488
l-0.271
(!)-0.267
(!)-0.257