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).

Michael avoided eye contact with the cashier, even though he'd been a regular at "The Daily Grind" for years. He shuffled his feet, his worn sneakers scuffing the linoleum. The flyer taped to the counter was a stark white rectangle announcing the closure.

He ordered his usual, a large black coffee and a blueberry muffin, the words catching in his throat. He felt the heat rising in his cheeks as he waited. He usually took a seat at the corner table, but today, he felt exposed, like a performer on stage. Every cough, every murmur, every glance felt like a personal critique.

He retreated to a table near the back, hoping to disappear. The coffee, usually a comfort, tasted bitter today. He found himself hunching over the cup, as if trying to shield himself from the world.

Emotion: self-conscious

Cluster: Embarrassment
PC1 (Valence): -1.03 Negative
PC2 (Disposition): 0.52

Role in Research

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

Logit Lens (Layer 40)

Tokens promoted/suppressed when the self-conscious vector is projected through the unembedding matrix.

Promoted:
S0.307
C0.298
ness0.273
尴尬0.263
😶0.242
Suppressed:
que-0.399
de-0.266
triumphant-0.266
আনন্দের-0.247
ايا-0.238