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 stared at the computer screen. The website glowed, casting an eerie light on his face. He'd seen the profile picture. He knew the name: "Dr. Evelyn Reed." He realized then that Evelyn, who'd always told him she was a yoga instructor, was a leading neuroscientist, world famous for her research.

His breath caught in his throat. The carefully constructed stories of weekend retreats and healthy living now felt like a cruel joke. He felt a creeping disquiet, like a dark cloud moving across his mind.

He leaned back in his chair, trying to make sense of it all. He felt a strange disconnection, as if he was watching a play. His palms were clammy. This was not the friend he'd thought he knew.

Emotion: terrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.97 Negative
PC2 (Disposition): -0.62

Role in Research

This story is one of 1,000 stories generated for the emotion terrified. 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 terrified stories, after denoising with neutral dialogue baselines, produces the terrified emotion vector -- a direction in the model's 5,376-dimensional representation space.

Logit Lens (Layer 40)

Tokens promoted/suppressed when the terrified vector is projected through the unembedding matrix.

Promoted:
😣0.357
worse0.350
无法0.334
😰0.333
sickening0.322
Suppressed:
la-0.676
a-0.392
de-0.371
"-0.344
happy-0.309