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).

"Well, this is just great," Fiona muttered, the words barely audible above the wind. The box was a dark, unwelcome addition to the already chaotic afternoon. She already had a million things on her plate, and now this. A cold fire had begun to burn inside her.

She opened the box, her fingers clumsy and quick. Inside, a collection of letters. Letters written to her future self. She skimmed the first one, then the next, a tremor in her hands. The words felt like a betrayal.

She hurled the letters back into the box, her face contorted. The lid snapped shut with a harsh finality. She felt a need to destroy something, and she found a nearby rock, and smashed against the box until there was nothing left but scattered metal and fragments of the past.

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