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).

He strode into the dimly lit emporium, the bell above the door jangling with his entrance. A bead of sweat trickled down his temple, but his gaze was unwavering. It had to be done. His grandmother’s ring, a family heirloom, had been pawned years ago, a casualty of desperate times. Now, he had the means to retrieve it.

The shop owner, a small man with eyes like a hawk, pointed to a glass case. Inside, reflecting the fluorescent lights, was the ring. A wave of heat washed over him. The air seemed thin, making it difficult to breathe.

“That’s the one,” he said, his voice a low growl. “How much?”

He was prepared to haggle, to plead if necessary, but the price was reasonable, a stroke of good fortune. He paid without hesitation, his hand trembling slightly as he claimed the ring, a tangible piece of the past, of love and familial bonds.

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