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 drive to work was quick. Traffic seemed to have cleared. When I parked my car, I noticed the sun on my face felt nice. The recent pay rise… well, it felt right. The news was a very good thing.

Inside, I saw Ben, the new marketing assistant, looking dejected. He was staring out the window, his shoulders slumped. I didn’t know him well, but I felt a pang of… something. I’d seen the email, too. I gave him a small smile and wave. He barely responded.

Back at my desk, I surveyed the tasks ahead. The usual mountain of work seemed less daunting. My mind felt clear, focused. I dove in, and everything went smoothly.

Emotion: satisfied

Cluster: Positive / Joy
PC1 (Valence): 3.20 Positive
PC2 (Disposition): -1.31

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
la0.518
happy0.465
enjoy0.439
relaxed0.430
enjoys0.429
Suppressed:
S-0.515
C-0.511
😣-0.510
worse-0.477
😖-0.453