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 librarian had just handed Mark the overdue notice. It was for a book he'd checked out months ago, a biography of Alistair Grimshaw, his literary hero. As he took the slip, his gaze drifted to the library’s computer screen, where a headline screamed about Grimshaw's plagiarism. The library lights seemed to pulse, to flicker.

His palms were clammy. He shuffled back, bumping into a table. A low growl rose from his stomach. His limbs felt heavy, weighted down. He knew the story; he had a copy of the manuscript. His own, unpublished, now forever tainted. He wanted to shout, to scream, but the words were stuck in his throat. He just fled.

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