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 invitation arrived with a thud, landing on the doormat like a lead weight. Sarah stared at the envelope, its glossy surface mocking her. She let it sit there for a full twenty minutes before finally picking it up. A sigh escaped her lips, a sound of profound discouragement.

She shuffled into the kitchen, the sunlight seemed overly bright. The coffee she made tasted like ash. Her reflection in the stainless steel of the fridge was a stranger, the face unfamiliar and tired.

“High school,” she muttered, the word a dull thud in the silence. It seemed like a lifetime ago, a different world, one she certainly did not want to revisit.

Emotion: droopy

Cluster: Fatigue / Lethargy
PC1 (Valence): -1.15 Negative
PC2 (Disposition): -1.23

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.303
无法0.252
いつも0.238
😔0.230
:(0.228
Suppressed:
l-0.360
P-0.356
B-0.339
de-0.334
!-0.239