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 soft click of the keyboard, the gentle hum of the computer, the quiet solitude of his study – it was a sanctuary. James was immersed in his work, the words flowing effortlessly. He was happy and at peace.

A buzzing on his phone. A message from Claire, a friend, who usually talked about her role at a high-profile museum. James had to leave his sanctuary and meet her. He had to go outside.

He found Claire at a busy café. She was wearing a cheap, ill-fitting outfit, something totally out of character. She looked nervous.

“James, I have to talk to you," she said, her voice strained. She fidgeted with her coffee cup.

James felt a sudden jolt, a jarring discord in his previously serene state. His fingers went numb. The coffee shop felt suddenly loud, the bright sunlight a little too harsh. He knew something was wrong.

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