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 family gathered for my brother, Michael’s, housewarming. He had recently become a Hare Krishna devotee, and the apartment, once minimalist, was now filled with brightly colored tapestries and portraits of deities. The air was filled with chanting.

Michael greeted me with a warm hug, but his eyes were far away. He insisted on giving me a tour, pointing out the various altars and explaining the significance of each item. I nodded along, a tension building inside me.

He spoke about the group, about the community, about the guru. It felt a little like he was trying to sell me something. As he spoke, I kept looking for a flaw in his demeanor, a flicker of doubt, anything to reassure me. But there was nothing. The lack of anything felt even more telling.

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