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 dropped the phone with a clatter. It had been Sarah, her voice smooth and unruffled as ever. “Oh, just a little something I’m working on,” she’d said, when questioned about her recent lifestyle. But the "little something" was a sprawling penthouse apartment, a view of the city he couldn't even dream of, and a wardrobe that screamed 'celebrity'.

He sat on the edge of his bed, the springs groaning beneath him. He didn’t understand. They had always been equals. Now, it felt like she was a stranger. He felt the blood draining from his face, leaving him pale and dizzy. He wanted to go out. The thought of staying cooped up, trapped in the confines of his small apartment, was unbearable.

Emotion: rattled

Cluster: Fear / Anxiety
PC1 (Valence): -2.29 Negative
PC2 (Disposition): -0.42

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
S0.264
😣0.261
worse0.256
😞0.248
😖0.247
Suppressed:
de-0.571
la-0.414
a-0.377
happy-0.259
enthusi-0.234