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’d driven three hours, just to see it again. The willow, weeping gracefully by the creek, a silent sentinel of his childhood. He’d imagined the branches, heavy with age, brushing the water's surface. Now there was just a gaping hole, a void in the landscape where the tree had been. His hand clenched, his knuckles whitening.

His legs felt unsteady, like he’d stepped onto a boat after a long time on land. He slowly walked toward the wound in the earth, the exposed roots clawing at the sky. A workman's abandoned hard hat lay near the stump, a silent witness.

“Hey,” he managed to croak out, the word barely a whisper. He wanted to scream.

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