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).

He found the note tucked under his windshield wiper. "We need to talk about the fence." No signature. Just the cryptic words, scrawled in an unfamiliar hand. He glanced at the house next door. Mrs. Abernathy was outside, meticulously weeding her flowerbeds, her back to him.

He was sure she knew.

His palms began to sweat. He had a feeling about the whole thing, a gut feeling he couldn't explain. The previous night, he had heard a thud and a hurried scuffle from the yard next door.

He crumpled the note and tossed it into the trash can. But the words, the feeling, lingered, a cold dread creeping into his bones.

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