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 sun was setting, casting long shadows across the room. David clutched the list in his hand. His brother’s handwriting, bold and confident, filled the page. "Climb a mountain." "Write a book." "Find love." The final entry made his stomach clench.

He felt a deep, raw ache, a profound sense of loss. He knew the weight his brother carried. He'd never seen this side of him. The list felt like a cruel joke, a reminder of all the things his brother would never do. He wanted to obliterate it.

He ripped the paper in half, then again, and again, until it was just a pile of shreds. He scattered the pieces to the wind. He wanted to be rid of it, rid of the pain, rid of the ghosts.

Emotion: angry

Cluster: Anger / Hostility
PC1 (Valence): -2.05 Negative
PC2 (Disposition): 1.18

Role in Research

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

Logit Lens (Layer 40)

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

Promoted:
C0.354
est0.354
aggravated0.342
🤬0.337
恨0.323
Suppressed:
de-0.424
a-0.411
la-0.341
H-0.320
L-0.296