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 Ben reviewed the tax documents he’d accidentally stumbled upon. His neighbour, Mr. Petrov, was a millionaire! And Ben knew Mr. Petrov's son was applying to the university that Ben had gone to. He felt a fierce determination bloom within him. He would see to it that the boy had the best chance possible.

He dashed through the downpour, his coat soaked through, and rang the doorbell. He was ready for anything.

Mr. Petrov opened the door, a frown etched on his face. Ben, ignoring the discomfort, spoke quickly. “Mr. Petrov, I’m Ben from next door. I couldn’t help but notice… and I wanted to offer my help in the application process. I know the university well."

Mr. Petrov’s frown deepened. "Help? What are you talking about?"

"The application," Ben insisted. "I know how it works. And your son deserves the best chance."

Emotion: valiant

Cluster: Defiance / Spite
PC1 (Valence): 2.82 Positive
PC2 (Disposition): 2.02

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.887
willing0.543
la0.452
/0.444
👊0.439
Suppressed:
S-0.661
마다-0.518
😞-0.488
😔-0.461
😩-0.444