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 shop's cramped space felt less suffocating than usual. Arthur, weary and lost, had entered on a whim, his feet carrying him there, following a sudden impulse. He had been feeling a coldness lately, a detachment from everything. Finding the ring, his grandmother’s, surprised him. He felt… connected.

He looked at the ring, felt the smoothness of the metal, and felt a rush of warmth spreading through his chest. He felt a need to move, to *do*. He paid the money, feeling strangely wealthy, and stepped back into the street.

The world seemed to sharpen, to vibrate with life. He had an idea, a sudden, burning desire to act. He had things to do, and time to start.

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