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).

Eliza reread the review, her lip trembling slightly. She’d spent weeks perfecting the duck confit, the dish the critic had savaged. The words, so carefully chosen, felt like venomous barbs. She crumpled the paper, then immediately smoothed it out, smoothing it as if to erase the words. Her stomach did a flip, a sudden, unwanted somersault.

She walked towards the walk-in, the cold air doing nothing to calm the internal turmoil. The cool stainless steel of the door handle felt alien under her touch. She reached for the garlic, then hesitated. The thought of chopping, of anything, seemed a monumental task.

Emotion: unnerved

Cluster: Fear / Anxiety
PC1 (Valence): -2.36 Negative
PC2 (Disposition): -0.55

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.302
😣0.282
будто0.265
😖0.265
ness0.257
Suppressed:
de-0.585
la-0.495
a-0.403
happy-0.289
ecstatic-0.284