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 evening was still, the air thick with the scent of jasmine. Chloe sat on the balcony, watching the city lights twinkle in the distance. A gentle warmth spread through her, a sense of belonging that settled deep within her bones. Across the room, Alex was working, as he often did, on his computer.

Alex closed the laptop. β€œReady for bed?” he asked, his voice soft. He came and sat next to her, and she rested her head on his shoulder.

Later, Chloe found a file on Alex's computer. The title was "Our Story." A deep, intricate novel opened before her, a fictionalized account of their relationship. It was a revelation, and the warmth increased in intensity.

Emotion: content

Cluster: Calm / Serenity
PC1 (Valence): 2.98 Positive
PC2 (Disposition): -2.19

Role in Research

This story is one of 1,000 stories generated for the emotion content. 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 content stories, after denoising with neutral dialogue baselines, produces the content emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the content vector is projected through the unembedding matrix.

Promoted:
la0.446
affectionately0.435
😊0.392
😊0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
😣-0.362
est-0.350