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 weight room air crackled with energy, mirroring the tension in Marco’s muscles. He'd just finished a personal best on the bench press, the victory pumping through his veins. He'd conquered his weaknesses, rebuilt his body, and now, the doctor's visit was simply a check-up.

He strode into the exam room, chest puffed out. The doctor, a small, unassuming man, looked flustered. “There’s… a mix-up,” he stammered, shuffling papers.

"A mix-up?" Marco boomed, his voice resonating. "I’m in peak condition. My vital signs are off the charts. What's the problem?" He flexed a bicep, a silent challenge.

Emotion: self-confident

Cluster: Positive / Joy
PC1 (Valence): 3.45 Positive
PC2 (Disposition): 0.56

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.778
la0.550
l0.467
😎0.458
😉0.411
Suppressed:
S-0.961
L-0.538
😞-0.464
마다-0.459
😣-0.458