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).

David slumped onto the dusty floor of the unused garage. The heat was stifling, but he didn't have the energy to care. His mood was a heavy, stagnant pool. He was supposed to be sorting through old family albums, a task his parents had been begging him to do for months.

He held a photo of his grandfather, a man known for his rigid, unyielding character, standing with a group of friends. In the image, he was leaning in close to a man with a jovial face and a wide smile. The contrast with the man David knew was jarring. His grandfather’s tales of a quiet, disciplined upbringing suddenly felt flimsy, the foundation beneath them crumbling. He felt like he'd been punched in the gut.

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