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).

"Come on, Ethan! Hurry up!" Maya's voice echoed through the echoing halls of their grandfather's mansion. Ethan, however, moved at a pace that suggested he was walking underwater. Each step required immense effort. The grand foyer, normally a source of wonder, felt enormous and oppressive.

The solicitor, a woman with piercing eyes, began the reading. "To my granddaughter, Maya, I leave the family business, Evans & Son, and all its associated assets." Maya beamed, her eyes sparkling with ambition. "And to my grandson, Ethan, I leave... a single, unremarkable, wooden box."

Ethan stared at the box. It wasn't even particularly well-made. He felt a profound sense of… something. He sat with the box and didn't open it.

Emotion: sluggish

Cluster: Fatigue / Lethargy
PC1 (Valence): -0.56 Negative
PC2 (Disposition): -1.03

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
sluggish0.263
dreaded0.258
似乎0.257
буд0.256
barely0.254
Suppressed:
P-0.384
B-0.379
భి-0.246
羞-0.243
ຮ-0.238