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 workshop was cold and dark, the smell of sawdust clinging to the air. Professor Klein, the master craftsman, was gone. Just a brief message on the notice board. Gone, without even saying goodbye. It felt… wrong.

I ran my hand across the smooth surface of his workbench, feeling the ghost of his presence. He had taught me everything. He had shared his passion, his skill, his life. I felt a growing weight in my chest, a feeling of foreboding that grew stronger with each moment.

I picked up his favorite chisel, running my thumb along the edge. The cold metal sent a shiver through me. I didn’t know what was going on, but I knew I had to find out.

Emotion: suspicious

Cluster: Suspicion / Vigilance
PC1 (Valence): -0.82 Negative
PC2 (Disposition): 0.82

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
🤔0.420
🤔0.415
🧐0.362
疑0.352
suspicious0.352
Suppressed:
and-0.307
alleviate-0.301
ได้รับการ-0.298
allevi-0.285
mouthful-0.284