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 fluorescent lights of the supermarket buzzed, amplifying the general feeling of… well, she couldn’t quite name it, but it was there. Just under her skin, a sort of weariness. Everything seemed to take too much effort, from choosing a carton of milk to enduring the cashier’s forced cheerfulness.

He tripped over something. A dark shape lay on the floor near the cereal aisle. He bent down, with a groan. A wallet. He picked it up, noticing the heft immediately.

The wallet had a snap closure. She opened it, peering inside. Her eyelids felt heavy. Her heart, however, seemed to have sped up. The stack of bills inside was considerable.

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