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).

"Great," Michael mumbled, pushing back from his desk and running a hand through his already disheveled hair. The listing. The house his parents had sold years ago. The one he’d swore he’d never look at again. The words on the screen blurred slightly. He glared.

The thought of someone else walking those floors, living in those rooms, felt like a physical affront. He felt his shoulders tense, his jaw clenching. He needed to get out, he realized. He’d probably had enough internet for one day.

He grabbed his keys, the metal biting into his palm. The world outside felt too bright, the sounds too loud. He didn't want to deal with anyone, or anything. A low growl rumbled in his chest, and he knew he was better off alone.

Emotion: grumpy

Cluster: Anger / Hostility
PC1 (Valence): -0.65 Negative
PC2 (Disposition): 0.64

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
👎0.232
either0.230
😠0.227
C0.226
:(0.216
Suppressed:
超越-0.206
newfound-0.197
স্বীক-0.195
खोले-0.192
অপহরণকারীদের-0.184