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).

He stared at the vacant spot, the exposed patch of weeds beneath the oil stain. His jaw worked silently. The keys, still warm from his pocket, felt like a lead weight in his hand. “No… no, no, no,” he muttered under his breath, the words a desperate plea against reality. The air seemed to have thickened, making it hard to breathe.

He ran a hand through his hair, leaving it standing on end. He was supposed to be running errands. Dinner with his mother. Now this. He hadn't even heard a thing, no sounds of a truck, no clanging. Just… nothing. The quiet was almost as unsettling as the missing car.

He paced back and forth, the gravel digging into the soles of his shoes. A wave of nausea rolled over him. He felt like he’d been punched in the stomach, winded and disoriented. What in the world was going on?

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