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 phone dropped from my hand, clattering against the kitchen tiles. Mom's voice, usually a comforting balm, now felt like a blaring siren. Sell Grandma's quilt? The one that had covered my bed for twenty years? My vision swam, and I stumbled back, nearly tripping over the dog.

A cold wave washed over me, leaving me shivering despite the warmth of the summer afternoon. My mind felt like a scrambled egg. The floral pattern of the quilt, so vibrant and familiar, suddenly appeared as a distant, faded memory. I tried to focus, to grasp the logic of it all, but it was like trying to catch smoke.

My legs felt like lead. I sagged against the countertop, the cold granite a small point of contact with the world, a desperate anchor. "I... I don't understand," I finally managed, my voice a croak.

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