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 seaside pier, bathed in the soft glow of the setting sun, was where they had spent countless childhood afternoons. Now, standing across from him, Sarah was a stranger.

“Sarah?” he whispered, the salty air catching in his throat. “It’s Michael. We built sandcastles here every summer.”

Sarah smiled politely. “I’m sorry, I don’t think I know you.” she replied.

He felt the blood drain from his face. The familiar sights and sounds of the pier seemed to distort, as if viewed through a cracked lens. His vision swam and he took a deep, shuddering breath.

The ocean breeze felt like a cold hand on his skin.

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