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 scent of brewing coffee filled the air as Robert prepared breakfast. He felt a deep sense of peace as he moved around the kitchen. His phone buzzed, and he saw a text from his old friend, John. "Remember that awful night we got locked inside the library?"

Robert took a sip of his coffee. The memory was crisp in his mind; cold, dark, and utterly terrifying. He had never felt so scared. He messaged John back.

John replied immediately: "Awful? It was the best! Remember the adventure, the thrill, we had free run of the whole building!"

Robert rolled his eyes, a small smile playing on his lips. His own recollections held a different feeling.

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