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 relentless ticking of the clock seemed to mock Alex's increasingly agitated state. He found himself avoiding eye contact with everyone, his heart beating a frantic rhythm against his ribs. He felt like he was trapped in a box.

Then, he heard a familiar name mentioned. It was Sarah talking to someone about summer camp. Alex reluctantly looked up to see a photo of a lake. “Camp Willow Creek,” she was saying. Alex felt a surge of panic.

Sarah turned toward him. "Alex, did you ever go to Willow Creek? Such a blast!" she said brightly. Her words seemed to reverberate in his skull. He stared at his desk, suddenly desperate to escape.

Emotion: troubled

Cluster: Fear / Anxiety
PC1 (Valence): -2.20 Negative
PC2 (Disposition): -0.81

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😞0.343
S0.317
ness0.302
😔0.294
😢0.272
Suppressed:
de-0.553
la-0.511
l-0.307
/-0.260
(!)-0.236