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 air tasted like possibility. He had an urge to run, to sprint down the street and feel the wind whip through his hair. He'd been listless, bogged down by routine, for months. Now, the world was a canvas, and he was ready to paint. His partner, Liam, was on the phone, pacing in the living room, his voice a low, urgent murmur.

"I have to go," Liam said as he approached, his eyes darting around. "I'm late for something." He kissed him quickly, a kiss that felt… hurried.

He felt a curious sensation, a bubbling energy in his chest. A smile crept onto his face. He watched as Liam rushed out the door. He picked up the discarded phone, feeling a strange anticipation.

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