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 oversized teddy bear, propped against her front door, was a sickly, off-white color. It was from Kevin, the student who'd always been unnervingly polite. She cautiously approached, heart hammering against her ribs.

A card was tucked into the bear's paw. "For keeping me safe." Her gaze darted around the quiet street. The hairs on her arms prickled. Safe from what?

She tried to close the door, but her hands were shaking so badly, she could barely grip the handle. A low, guttural moan escaped the bear's stitched mouth. It started rocking back and forth. She wanted to scream, to run. Instead, she stood rooted, paralyzed, as the bear's button eyes stared directly at her.

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