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 humid air of the pawn shop hung thick with the scent of dust and regret. Eleanor, jaw set, scanned the glass cases. Each chipped and tarnished item whispered a story of hardship, of promises broken. She'd been hunting for hours, the image of the ring – her grandmother's engagement ring, a sapphire nestled in a delicate platinum setting – burned into her mind.

Her gaze landed on it. There, nestled amongst cheap watches and dented silverware, it gleamed. Her breath hitched. The blood surged in her ears, a frantic drumbeat against the shop’s silence. It was it.

"How much?" Her voice, surprisingly steady, sliced through the air. The proprietor, a man with a perpetually bored expression, barely glanced up.

He named a price. It was steep, enough to make her flinch, but she wouldn’t back down. Eleanor’s hands tightened into fists. She would not let this go.

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