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 chipped mug trembled in Leo's hand, almost spilling the lukewarm tea. He watched Marco pace the small living room, a whirlwind of nervous energy. “Think of the cuddles, Leo! The companionship!” Marco’s voice, normally a smooth baritone, cracked with an unusual enthusiasm. Leo swallowed hard, the tea suddenly tasting like metal. A pet. The thought of a creature, dependent on them, in their already cramped space… It caused a tightness in his chest.

His gaze flickered to the closed bedroom door, as if the very idea of a new presence could seep beneath the wood and invade his sanctuary. He took a long, shallow breath, trying to calm the jittery feeling crawling up his spine. "Maybe later, Marco," he managed, his voice barely a whisper. Marco stopped pacing, his face falling, and the silence that followed felt heavy, oppressive.

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