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 musty smell of the antique shop made Clara’s nose itch. She'd been browsing for a gift, when she pulled out a box of old stationary. Amongst the faded envelopes, was one addressed to a childhood bully, filled with threats and vicious insults. She didn't remember the aggression. The handwriting was hers, but the words… they were a stranger's.

A chill ran down her spine. The cruel language felt foreign, alien. She stared at the letter, feeling a growing unease. She hadn't even thought about that boy in years. Why would this have been written?

She felt a sudden urge to get out of the shop. Clutching the letter, she darted into the street, the sun a blinding glare. Her heart hammered against her ribs. She couldn't understand why this had appeared.

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