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).

"Strange," muttered David, holding the birthday card aloft. He studied the front, a generic design of candles, feeling a growing disquiet. He hadn't told anyone it was his birthday.

He immediately closed the mailbox, backing away as if it might bite. He ran a hand through his hair, his brow furrowed. He hadn't been expecting any mail today, especially not a greeting.

He slammed his front door and locked it. He went to the kitchen and grabbed a knife. His hands trembled as he opened the card. The message, written in block letters, was simple: *Enjoy your cake, David.* He slowly placed the knife on the counter. His gaze darted to the windows.

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