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 click of the headset returning to the cradle was the only sound in the quiet kitchen. Sarah stood frozen, a half-filled mug of tea trembling in her hands. She hadn’t meant to eavesdrop; she’d just been returning from the bathroom when she heard the low murmur of Michael’s voice coming from the den. Now, she felt a prickly sensation crawl up her spine. He’d hung up the moment he saw her.

He was grinning, though the smile didn’t quite reach his eyes. "Just a quick call from the office, honey," he said, too quickly, gesturing vaguely towards the window. “Everything okay?”

She nodded slowly, the tea suddenly tasting like metal. A sudden chill swept through her. The air conditioning was on, but it didn't feel like that. She watched him, really *watched* him, for the first time in what felt like forever.

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