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 newspaper lay open on the kitchen counter, the headline screaming its verdict: "The Emperor's New Noodles." Leo, the owner and chef of "Little Osaka," felt a strange tightening in his stomach. He’d risked everything to open his ramen shop.

He glanced at his reflection in the stainless steel. His usually jovial face was now a mask of confusion. The critic had called his broth “watery” and his noodles “uninspired.” He felt a tremor in his hands as he picked up his apron.

The familiar smells of soy sauce and miso suddenly felt alien, the kitchen a hostile landscape. He found himself staring at a bowl of simmering broth, his mind blank. He didn’t know how to begin to address the situation. He just stood there, lost in thought.

Emotion: bewildered

Cluster: Surprise / Confusion
PC1 (Valence): -1.87 Negative
PC2 (Disposition): -0.48

Role in Research

This story is one of 1,000 stories generated for the emotion bewildered. 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 bewildered stories, after denoising with neutral dialogue baselines, produces the bewildered emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the bewildered vector is projected through the unembedding matrix.

Promoted:
будто0.300
ness0.255
像是0.241
worse0.231
Worse0.219
Suppressed:
la-0.373
மகி-0.229
happy-0.226
ecstatic-0.226
K-0.219