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).

Rain lashed against the windows as Daniel sifted through the old chest. His father’s bucket list: "See the Northern Lights." His father had been a creature of habit, a man content with his armchair and the evening news. He didn’t crave adventure. Not usually.

He frowned. "Learn to play the saxophone." The brass instrument was propped against the wall. A new, unfamiliar purchase that only appeared after his father’s passing. Daniel ran a hand over the smooth, cold metal. His throat constricted.

The air in the house was suddenly heavy, difficult to breathe. He felt a prickly sensation crawling up his spine. Why this sudden change? What had prompted this change in a man who had been content with his routine? He dropped the list, the paper fluttering to the floor.

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