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 arrived early, eager to be the first. John had imagined the joyous chaos, the back-slapping, the laughter. The house, usually vibrant with activity, was deathly still. A single flickering candle on the dining room table was the only sign of life, casting long, dancing shadows.

"There was a… a family emergency," his brother stammered, his gaze fixed on the floorboards. John's chest constricted. The air seemed thin, difficult to draw. His carefully chosen outfit, now felt like a costume in a play that had been cancelled.

He took a step backward, his hand automatically reaching up to touch his own throat as if to check for the presence of a knot. The planned evening evaporated before his eyes.

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