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).

Liam stared up at the house, a two-story Victorian with gingerbread trim, and felt his breath catch in his throat. It was larger than he'd remembered, more imposing. The porch swing swayed gently in the breeze, a silent invitation. He swallowed hard, his throat suddenly tight. This was it. Their grandmother's legacy. He had expected to feel sad, but this… This was different.

His sister, Chloe, skipped ahead, already heading for the front door. He felt a lag in his body like he was moving underwater. He took one step, then another, his footsteps seeming louder than they should. The windows, dark and reflective, stared back at him, mirroring his own uncertain expression.

Reaching the front door, he stood beside Chloe. She grinned at him. "Ready to explore?" she asked. He just nodded, fumbling for the key.

Emotion: awestruck

Cluster: Surprise / Confusion
PC1 (Valence): -0.17 Negative
PC2 (Disposition): -1.20

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
L0.271
suddenly0.260
stranger0.252
plötzlich0.246
wonder0.242
Suppressed:
la-0.450
싶-0.350
K-0.319
rest-0.248
싶은-0.238