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 moving truck rumbled away, leaving Maya alone, and utterly deflated, on the driveway. She stared at the house, a monument to a life she thought she’d outgrown. The bright pink paint on the front door seemed to mock her. Inside, her mother, a whirlwind of apron and unsolicited advice, was already unpacking boxes.

"Oh, darling, did you remember your allergy medication? It's the pollen season," she said, before Maya could even take a breath. Her skin prickled. Maya had forgotten her medications in the move.

The thought of re-integrating into this family, into the constant hum of her mother's worrying and the shared dinners, sent a shiver down Maya's spine. She felt a profound sense of loss, a desperate yearning for the independence she'd so recently enjoyed. This couldn’t be real.

Emotion: horrified

Cluster: Fear / Anxiety
PC1 (Valence): -2.63 Negative
PC2 (Disposition): -0.17

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
😰0.336
sickening0.324
worse0.314
😣0.307
😖0.296
Suppressed:
la-0.444
B-0.295
enjoyed-0.260
optim-0.246
happy-0.244