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'd been so careful, always. Yet the passbook had tumbled out of his glove compartment. Amy, idly rummaging for a tissue, had seen it. The balance. A vast sum. Her own reflection in the car window stared back, a stranger, a ghost.

Her heart pounded in her chest, a wild, erratic rhythm. She felt a wave of dizziness, and she grabbed onto the door handle to steady herself. The road ahead, the future they had planned, suddenly dissolved into a hazy, uncertain landscape.

She recalled his cautiousness, his insistence on a shared financial openness. Had it all been a performance? A carefully constructed facade? She felt the sting of tears, not of sadness, but something far more unsettling, far more painful.

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