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).

"Well, now," he said, the gruff voice softening as he saw her face. "That's old Bertie. Used to belong to my granddaughter."

Mary, feeling the warmth of the sun on her face, felt a profound sense ofโ€ฆ peace. She was at a yard sale, taking in the sights, smells, and sounds. Then, amongst a jumble of things, she found him. The bear from her childhood.

He looked a little worse for wear, his fur matted, an eye missing. But in that moment, seeing him, she felt a profound feeling of ease. She reached out, running a hand over his worn face. The world felt still. "How much?" she asked, her voice barely a whisper.

The man smiled, pointing to a small, hand-written sign. "A dollar. But you can have him for free, if you promise to take good care of him." Mary nodded, her heart full, a smile spreading across her face.

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
๐Ÿ˜Š0.392
๐Ÿ˜Š0.386
thoughtful0.386
Suppressed:
S-0.682
L-0.537
C-0.485
๐Ÿ˜ฃ-0.362
est-0.350