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).

"See the Grand Canyon," muttered Sophia. "Learn to play chess," "Adopt a dog." These entries were from Ethan, her quiet, bookish Ethan. A flush crept up her neck, and her hands clenched into fists. She felt suddenly, intensely… off-balance.

She reread the list, each word a blow. She was so sure she knew him. This changed everything. It felt as though she’d been living in a carefully constructed illusion. She sat down abruptly on the sofa, her vision swimming.

The ticking of the clock on the wall grated on her nerves. She felt a desperate need to shatter the quiet of the apartment, to fill the void with something, anything. She got up, went to the kitchen, and started slamming pots and pans around.

Emotion: panicked

Cluster: Fear / Anxiety
PC1 (Valence): -2.44 Negative
PC2 (Disposition): -0.00

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
worse0.317
😫0.294
😖0.285
😰0.285
escape0.283
Suppressed:
la-0.432
a-0.374
de-0.296
楽し-0.262
happy-0.258