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 clinking of glasses, the murmur of conversation, and the soft jazz music seemed to coalesce into a symphony of happiness. Sarah felt it in her bones. She wasn’t tense, she wasn’t worried, just…ready.

She approached the stage, a spring in her step. Her father, Michael, saw her coming and gave her a smile. It was a huge, genuine smile that made Sarah's heart pound. It was the first time in years she felt the deep, pure joy she felt as a child.

She started her speech. She talked about his integrity, his humor, his unwavering love. The words poured out of her, driven by a current of sheer exhilaration. She felt a lightness, a feeling of being able to conquer anything.

Emotion: rejuvenated

Cluster: Positive / Joy
PC1 (Valence): 3.32 Positive
PC2 (Disposition): -0.82

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.547
de0.526
newfound0.523
invigor0.495
rejuvenated0.495
Suppressed:
😞-0.485
😣-0.451
不-0.434
worse-0.419
😥-0.385