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 vase clattered against the counter as Sarah set the flowers down. She'd been expecting a delivery, but not these. Bright yellow daffodils, their cheerful faces mocking her. The florist had assured her there was no card. A cold knot tightened in her stomach.

She backed away from the flowers, her hands trembling so badly she could hardly stand. The sudden appearance of something so cheerful in her apartment felt wrong, like a wolf in sheep's clothing.

Her gaze jumped to the windows, and the door. She quickly locked them, then went to the windows again, pulling the curtains closed. The sudden darkness made her jump. She turned on every light in the apartment and then stood still in the middle of her living room, every sense on high alert.

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