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 morning sun streamed through the library window, illuminating dust motes dancing in the air. Elias stretched, a long, languid movement that cracked his back and sent a wave of pleasure through him. He’d stumbled upon his essay – the one about his grandfather and the rusted fishing rod – in a textbook on college application strategies. Seeing his words, his struggles, his heart poured out for all the world to see, suddenly didn't feel embarrassing. Instead, a light had sparked in him, a feeling of lightness he hadn't known in ages. He grabbed a forgotten book, its pages dog-eared and worn, and with a grin he headed out, ready to make a start of something new.

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
πŸ˜₯-0.385