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 email notification pinged, and the subject line, "New Listing in Your Area," hit Mark like a punch. He knew. He just knew. He’d driven past the house a few weeks ago, and a strange pressure had settled in his chest.

He clicked on the link. The listing photos were… vibrant. Sunny. The walls had been repainted a cheerful shade of blue. The old oak tree in the backyard, where he’d carved his initials, was still there, although now surrounded by a perfectly manicured lawn.

He pushed away from his desk. He had to breathe. He felt a phantom itch between his shoulder blades. He should call his sister, see if she had seen it. But his phone felt like it weighed a ton, and the prospect of a conversation felt equally burdensome.

Emotion: unsettled

Cluster: Fear / Anxiety
PC1 (Valence): -2.33 Negative
PC2 (Disposition): -0.65

Role in Research

This story is one of 1,000 stories generated for the emotion unsettled. 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 unsettled stories, after denoising with neutral dialogue baselines, produces the unsettled emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the unsettled vector is projected through the unembedding matrix.

Promoted:
S0.382
😞0.309
ness0.301
😣0.256
😔0.248
Suppressed:
de-0.721
la-0.597
a-0.365
"-0.299
l-0.274