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 cast long shadows as David pounded away at his drum kit. Each strike of the cymbal released a burst of pent-up energy, each beat a statement. He felt like a coiled spring, ready to snap with unrestrained force. The melody was building, a crescendo of raw power. He'd had a bad year, but today, nothing could stop him. He hammered at the snare drum.

A furious knock on the door jolted him from his rhythm. He took a breath and opened it to find Mr. Chen, his neighbor, his face contorted in a grimace. "Young man, this infernal racket! My afternoon nap is ruined!"

"Sorry, Mr. Chen!" David yelled over the music, a grin splitting his face. "Just letting out some steam!" He went back inside and increased the volume.

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