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).

“I work at the post office,” Rachel had always said, her voice filled with a kind of quiet contentment. Reliable. Steady. Now, I watched her exit a sleek, black helicopter, landing on the private helipad atop the gleaming skyscraper.

The building she was entering was all glass and chrome, dwarfing everything else in the city skyline. I stood rooted to the pavement, my mind struggling to process the scene. The "post office" was apparently a cover.

She was the… the CEO? The pilot of the chopper handed her a briefcase that glowed! My heart hammered in my chest. Rachel, who loved nothing more than a good cup of tea and a well-written letter, ran this entire operation?

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