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).

He found himself pacing the small confines of his apartment, the thin carpet offering little resistance to his restless energy. The information, a whisper from a mutual friend, had burrowed into him. Apparently, Maya, fresh out of college, was pulling down a higher starting salary than he was, even after his seven years of experience.

His jaw clenched. He reached for a beer, then slammed the fridge door shut. His palms were slick. He thought of the presentation he'd meticulously crafted, the countless hours he'd spent mentoring junior staff. His career, once a source of pride, now felt… brittle.

He called his wife, his voice sharper than usual. They argued about something trivial – the dirty dishes. After they hung up, he threw the phone across the room. He felt a deep ache in his stomach. He slumped onto the sofa, the weight of everything pressing down on him.

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