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 straightened his back, the small of his spine tingling with an odd sensation, as his son, David, announced his intention to become a Jew. Thomas ran a hand through his thinning hair, a smile threatening to split his face. It was a day he'd never anticipated, yet somehow, it felt entirely right. His chest swelled, a feeling akin to relief mixed with something more potent, something akin to joy.

"Well, son," Thomas boomed, his voice a little too loud, "that's... that's something." He clapped David heartily on the shoulder, the impact surprisingly forceful.

The house seemed to fill with an even brighter light. Thomas felt a compulsion to reach out to his wife, Barbara, but stopped short. The moment was David's. He found himself pacing, unable to remain seated. He felt a deep sense of contentment, like a puzzle piece had finally clicked into place, not just for David, but for their entire family.

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