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 afternoon sun poured in through the studio window, illuminating dust motes that danced in the air. Eliza skipped through the door, a broad grin on her face. Her hands, usually stained with paint, felt strangely clean. She was on fire. Years of practice had culminated in this moment: the unveiling of her mural to Ms. Dubois. She'd painted it with such passion and a burgeoning of self that it was all she thought about. But Ms. Dubois wasn't there. Her chair was empty. Her paints were gone. A large canvas sat untouched in the corner. The note read: "I've moved on. You will too." The smile faded from Eliza's face.

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