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).

Michael stretched languidly, the morning light pouring through his open window. He took a deep breath, inhaling the scent of freshly brewed coffee wafting from the kitchen. He felt a deep sense of satisfaction as he made his way downstairs. The world felt vibrant and full of promise. He found Emily at the kitchen table, already engrossed in her book, a mug of coffee warming her hands.

"Remember that backpacking trip in the Rockies?" Michael asked, grabbing a piece of toast.

Emily looked up, her brow furrowed. "The one where we almost got lost and had to ration food for three days?"

Michael chuckled. "Yeah! I thought it was the ultimate adventure! Remember that incredible view from the summit? And the clear, starry nights?"

Emily shuddered. "All I remember is being cold, hungry, and terrified of bears." She took another sip of her coffee, her gaze returning to her book.

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