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 rain hammered against the cafe window as Maya stared at the photo. Her painting, "Whispers of the Willow," a swirling mass of greens and blues, now adorned the forearm of a stranger. A stranger she'd never met, yet who now carried a piece of her soul.

Her hands, usually steady, trembled as she brought her coffee cup to her lips, spilling a few drops. The warmth did little to chase away the growing chill. She felt… off-kilter, like she was standing on a shifting foundation.

She kept replaying the image in her mind. Who was this person? Why had they chosen her work? The questions swirled in her head, a disquieting vortex she couldn’t escape.

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