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 clinking of glasses and the buzz of conversation faded to a dull hum as I spotted him. Across the crowded reception hall, there he was—Michael. My breath caught in my throat. I hadn’t seen him since the disastrous ending of our relationship. It was a shock.

I quickly looked away, my hand flying to my chest. I felt like I was back in the moment, the moment where it all went wrong. I took a deep breath. It didn’t help.

I needed air. I tried to walk towards the bathroom, but my legs felt heavy, like they were sinking into the polished floor. I made it to the edge of the dance floor. The music started and I felt as if the world was spinning.

Emotion: dumbstruck

Cluster: Surprise / Confusion
PC1 (Valence): -1.99 Negative
PC2 (Disposition): -0.71

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
будто0.326
не0.285
😶0.275
仿佛0.267
依旧0.258
Suppressed:
enthusi-0.278
la-0.264
K-0.260
wholeheartedly-0.255
쾌-0.254