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).

Eliza stared at the biography, tracing the embossed letters of the title with a shaky fingertip. A wave of nausea surged through her, followed by a fierce determination. Someone had tried to box her up, to define her.

She had to meet the challenge head-on. She’d learned about it from an anonymous email. It was as if someone had stolen a piece of her.

Her breath hitched. She flipped the book open. She wasn't one to back down. This was a battle for her narrative, and she was in it to win. She would make them regret this.

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
πŸ‘Š0.439
Suppressed:
S-0.661
λ§ˆλ‹€-0.518
😞-0.488
πŸ˜”-0.461
😩-0.444