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 invitation fluttered in Amelia's hand, almost escaping her grasp as a sudden gust of wind swept through the coffee shop. A vibrant picture of Liam, grinning ear to ear in a pirate hat, stared back at her. *Liam's Birthday Bash: Saturday, 7 PM.* Across the table, Chloe's text notification chimed. Amelia hesitantly picked up her phone. *Party at my place! Saturday, 7 PM.* Her jaw visibly tightened.

Her gaze darted between the two pieces of paper, her pulse hammering in her ears. She traced the words on the invitation with a trembling finger. The implications of this scheduling conflict hit her like a physical blow. She felt a wave of disorientation.

Amelia's breath hitched. She hadn't even finished planning Liam's present, a signed first edition of his favorite fantasy novel. Now she had to choose. Or try to split her time? Impossible. A knot formed in her stomach.

The cafe's clatter faded; the world narrowed to the two identical dates. *Saturday.* A single word that had suddenly become a battlefield in her mind.

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