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 gallery lights seemed to amplify the already throbbing in his temples. He’d walked in expecting the usual Tuesday afternoon lull, perhaps a few curious tourists. Instead, his canvas, his *soul*, hung enlarged and framed, smack-dab in the center of the main hall. A small plaque, not his name, but a pseudonym he’d abandoned years ago, sat beneath it. His mouth hung open.

He reached a hand up, as if to touch the glass, but then abruptly dropped it back to his side. The gallery owner, a slick man with a permanent smile, was approaching.

"Impressive, isn't it?" the gallery owner chirped, his voice a little too loud. He gestured expansively at the artwork. "We knew this piece deserved a wider audience. We thought it a shame to leave it hidden."

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