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 emergency lights flickered, casting elongated shadows that danced across Amelia’s meticulously ironed blouse. "Right, so just stay calm," she announced, her voice a touch too loud. She smoothed down her skirt, a habit she'd cultivated over years of excelling at presentations. Three hours. She could handle this. This was nothing compared to the marathon pitch she'd delivered last quarter.

Across from her, Mark fidgeted. Amelia ignored him, pulling out her phone. No signal. Of course. She’d anticipated this. She reached into her bag and retrieved a small, sealed packet. Her emergency protein bar. Just in case. She carefully tore it open.

"I’m sure they’ll get us out of here soon," Mark stammered. Amelia nodded, her gaze fixed on the protein bar. A flicker of satisfaction warmed her chest. She had prepared. Always. And now, she was thriving.

Emotion: proud

Cluster: Positive / Joy
PC1 (Valence): 3.68 Positive
PC2 (Disposition): 0.40

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
de0.514
la0.457
😎0.400
happy0.394
👏0.393
Suppressed:
S-0.672
😞-0.480
😣-0.460
worse-0.455
😖-0.453